|88| Anti harassment bullying task force materials (combined)

Brooke R. Whitted Whitted, Cleary + Takiff, LLC 3000 Dundee Road, Suite 303

Northbrook, IL 60062 Phone: (847) 563-8662

Fax: (847) 564-8419 [email protected]

www.wct-law.com

September 20, 2010 December 13, 2010

ANTI-HARASSMENT/

BULLYING TASK FORCE

MEETING AGENDA AND

HANDOUT

mailto:[email protected]�

http://www.wct-law.com/�

1

Compiled by: Brooke R. Whitted

Whitted, Cleary & Takiff, LLC 3000 Dundee Rd, Suite 303 Northbrook, Illinois 60062

Ph. 847-564-8662 Fax 847-564-8419

www.wct-law.com [email protected]

ANTI HARASSMENT/ BULLYING TASK FORCE

Scheduled Meetings



2

MEMO To: Friends From: BRW Date: 12/07/10 Re: Next Meetings

Please be advised that the next Task Force meeting is Monday, December 13th from 3:00 p.m. to 6:00 p.m. The meeting will occur simultaneously in Springfield and Chicago at the State Board of Education V-Tel rooms in each of the State Board’s respective buildings:

In Chicago: The State of Illinois Building, 14th floor, Illinois State Board of Education V-Tel room. In Springfield: The Alzina Building, 100 N. First Street, 3rd floor, V-Tel room.

These meetings are public meetings and therefore governed by the Open Meetings Act. The convener is Mr. Darren Reisberg, who is the General Counsel for the Illinois State Board of Education. Minutes from the last meeting are available on the corresponding website maintained by the Illinois State Board of Education, at this link: http://www.isbe.net/SBPTF/default.htm. In addition to the next meeting on December 13th, there are two more scheduled at the same locations: January 10, 2011 from 3:00 – 6:00 p.m. and February 14, 2011 from 3:00 – 6:00 p.m. This meeting time was decided upon by the Task Force as an accommodation to the students on the Task Force who have to be in school.

http://www.isbe.net/SBPTF/default.htm�

1



Ph. 847-564-8662 Fax 847-564-8419



Task Force Membership



2

Task Force Membership Attached please find a document listing the members of the Anti-Harassment/Bullying Task Force created by statute. The link to the ISBE web page, containing minutes and future meetings is at http://www.isbe.net/SBPTF/default.htm.

http://www.isbe.net/SBPTF/default.htm�

1



Ph. 847-564-8662 Fax 847-564-8419



Notes #1



2

Bullying Task Force Notes Brooke R. Whitted

According to Sharon Hirsch, MD, Director of Child & Adolescent Psychiatry of the Department of Psychiatry at the University of Chicago, the limiting factor in eliminating bullying is the lack of availability of mental health services, and services for abused children. According to Dr. Hirsch, children who have been abused are more likely to turn into abusers (potential bullies) themselves and it is critical that we have intervention programs for this. The availability of good mentors is also important, as is effective after school programming. According to Ken G. Papineau, Director of Coordinated School Heath/Office of Special Education and Supports of the Chicago Public Schools, Task Force members should become familiar with the national, state and localized data available concerning health risk behaviors as collected through the Youth Risk Behavior survey, a longitudinal survey conducted every two years, and administered by the Center for Disease Control and Prevention (CDC). The public may obtain data related to the Youth Risk Behavior survey from the CDC by visiting: http://www.cdc.gov/healthyyouth/yrbs/index.htm. This survey provides school districts with valuable student reported data regarding the Nation’s Leading Health Indicators (“LHIs”), including the following measures related to school violence:

1. Carried a weapon on school property on at least one day;

2. Did not go to school because they felt unsafe at school or on their way to or from school on at least one day;

3. Threatened or injured with a weapon on school property one or more times;

4. In a physical fight on school property one or more times;

5. Bullied on school property.

Mr. Papineau also suggests that the Illinois State Board of Education’s School Bullying Prevention Task Force discuss their goals and objectives with the Illinois Department of Public School Health Risk Task Force, as both committees have been established by the Governor’s office and may have complimentary goals and objectives. Mr. Papineau can be reached at 773/553-1830. Attached please also find an op-ed on bullying written by Dr. David Fassler. Additionally, Dr. Fassler provides an attached list of links to relevant articles on bullying. Finally, Yvonne E. Muirhead, Research Coordinator for the FBI’s Behavioral Analysis Unit, sent contact information for Dr. Dudley Cornell at the University of Virginia, an expert

http://psychiatry.uchicago.edu/�

http://www.cps.edu/Programs/Pathways_to_success/SpecialEducationandSupports/Pages/OfficeofSpecialEducationandSupports.aspx�

http://www.cps.edu/Programs/Pathways_to_success/SpecialEducationandSupports/Pages/OfficeofSpecialEducationandSupports.aspx�

http://www.cdc.gov/�

http://www.cdc.gov/�

http://www.cdc.gov/healthyyouth/yrbs/index.htm�

http://www.fbi.gov/about-us/training/bsu�

http://www.fbi.gov/about-us/training/bsu�

3

on whom the FBI relies in situations relative to harassment and bullying. Clearly such cases involving the FBI concern school shootings. The document she sent is also attached.

1

The Finnish KiVa System



Ph. 847-564-8662 Fax 847-564-8419



Notes #2



2

The Finnish System for Harassment Prevention Brooke R. Whitted

The Finnish Ministry of Education has attracted attention worldwide as a result of their well established fifty percent reduction in bullying and harassment in their school system. The program they have developed is called KiVA and this program concentrates on the onlookers, or bystanders, and enlisting their help to support a victim. Part of the requirement of the KiVA program is an introduction into the curriculum of a minimum of twenty hours per year of lessons involving discussion, group work, films, and hands on role playing. There is also an anti-bullying computer game and a multitude of other virtual reality materials. The KiVa system also has established some low tech but highly effective procedures including extremely high visibility vests for teachers supervising recess activities; posters all over the school buildings; a PowerPoint for parents, and a Parent Guide including advice on how to reduce bullying behaviors. For acute cases, there is established a team of three teachers in each school. Meetings with classmates are conducted regularly to encourage them to support victims. There is a manual that gives structure to and guides discussions with students and staff, and action plans are developed which are research based and formalized system wide to eliminate any and all atmospheres in a particular building that encourage casual violence. There is an evaluation now being done with a total number of students studied of thirty thousand, two hundred and thirty-four schools, and one thousand classrooms. Preliminary evidence is as follows:

1. The average rate across all countries of bullying and harassment is thirty-three percent;

2. With regard to effectiveness studies, these include in part:

A. Whitted and Duppr (2005), Children and Schools, Volume 27 (3) pages 167-175;

http://www.minedu.fi/OPM/?lang=en�

http://www.kivakoulu.fi/content/view/56/171/�

3

B. Smith, Cousins and Stewart, Ingredients of Effective Programs, Canadian Journal of Education, Volume 28 (4) pages 739-762;

C. Smith, Peppler and Rigby, Bullying In Schools: How Successful Can Interventions Be?, New York, NY Cambridge University Press;

D. Tutty, LM: Best Practices in School – Base Bullying Prevention Programs: What Works? Prevnet Volume 1, pages 144-165, Kingston, Canada Publisher, Prevnet Incorporated.

Under the KiVA system, effective approaches must be multidisciplinary; involve parents; be system wide; be across the curriculum and seamless; utilize relationships; keep bullying issues front and center and not just the subject of one isolated meeting. The KiVa approach also enlists the creative use of high status peers in encouraging bystanders to speak up in support of victims. In addition, it was well established that the entire program has to be implemented and not just parts of it, and attention has to be shifted away from those who bully and support given to those who are victimized.

1

STATISTICS



Ph. 847-564-8662 Fax 847-564-8419



Notes #3



2

Bullying/Harassment Notes

1. Only one-third of all incidents are reported to a teacher (page 3, REL Report). 2. Gender based harassment is less likely to be repeated. 3. Definition: Exposure, repeated and over time, to negative actions on the part of one or more other students.

BRW feels this definition is somewhat limited.

4. Underreporting hampers ability of educators to respond, but what about the dismissal of a bona fide report? Is this a problem with the definition? 5. Making fun; spreading rumors; social exclusion all tend not to be reported. 6. About thirty to thirty-two percent of all students are bullied in some form each year. 7. Many school districts have set up an anonymous hotline reporting system to encourage more reporting. BRW note: This should be mandated in all schools. 8. One question asked in research was “Was a teacher or some other adult notified about the event?” The weakness of this is that there was no inquiry about the response and then lack of adult response being a direct cause of non-reporting. BRW Note: Tracking of responses should be mandated. 9. Statistics: Injury = 60% reported Threat = 55% reported Property destruction = 51% reported Pushing/shoving/tripping = 46% reported Once/twice per week = 44% reported

http://ies.ed.gov/ncee/edlabs/regions/northeast/pdf/REL_2010092.pdf�

3

Everyday = 48% percent reported Sixth grade: 53% percent reported Twelfth grade: 27% percent reported 10. There are no differences with respect to reporting across racial or gender groups. 11. There is no relation, for reporting, to a student’s performance in school.

I S S U E S & A N S W E R S R E L 2 0 1 0 – N o . 0 9 2

At Education Development Center, Inc.

What characteristics of bullying, bullying victims, and schools are associated with increased reporting of bullying to school of ficials?

I S S U E S&ANSWERS R E L 2 0 1 0 – N o . 0 9 2

At Education Development Center, Inc.

What characteristics of bullying, bullying victims, and schools are

associated with increased reporting of bullying to school officials?

August 2010

Prepared by

Anthony PetrosinoLearning Innovations at WestEd

Sarah GuckenburgLearning Innovations at WestEd

Jill DeVoeAmerican Institutes of Research

Thomas HansonWestEd and Regional Educational Laboratory West

WA

OR

ID

MT

NV

CA

UT

AZ

WY

ND

SD

NE

KS CO

NM

TX

OK

CO

AR

LA

MS AL GA

SC

NC

VA WV

KY

TN

PA

NY

FL

AK

MN

WI

IA

IL IN

MI

OH

VT

NH

ME

CT RI

MA

MO

VI

PR At Education Development

Center, Inc.

Issues & Answers is an ongoing series of reports from short-term Fast Response Projects conducted by the regional educational laboratories on current education issues of importance at local, state, and regional levels. Fast Response Project topics change to reflect new issues, as identified through lab outreach and requests for assistance from policymakers and educators at state and local levels and from communities, businesses, parents, families, and youth. All Issues & Answers reports meet Institute of Education Sciences standards for scientifically valid research.

August 2010

This report was prepared for the Institute of Education Sciences (IES) under Contract ED-06-CO-0025 by Regional Educational Laboratory Northeast administered by Education Development Center, Inc. The content of the publication does not necessarily reflect the views or policies of IES or the U.S. Department of Education nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

This report is in the public domain. While permission to reprint this publication is not necessary, it should be cited as:

Petrosino, A., Guckenburg, S., DeVoe, J., and Hanson, T. (2010). What characteristics of bullying, bullying victims, and schools are associated with increased reporting of bullying to school officials? (Issues & Answers Report, REL 2010–No. 092). Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Educational Laboratory Northeast and Islands. Retrieved from http://ies.ed.gov/ncee/ edlabs.

This report is available on the regional educational laboratory web site at http://ies.ed.gov/ncee/edlabs.

http://ies.ed.gov/ncee/edlabs



Summary REL 2010–No. 092

What characteristics of bullying, bullying victims, and schools are associated with increased reporting of bullying to school officials?

This study tested 51 characteristics of bullying victimization, bullying victims, and bullying victims’ schools to deter-mine which were associated with report-ing to school officials. It found that 11 characteristics in two categories—bully-ing victimization and bullying victims—showed a statistically significant asso-ciation with reporting. The study also notes the high percentage (64 percent) of respondents who experienced bullying but did not report it.

Bullying appears to be frequent among U.S. students and has been associated with several short- and long-term negative consequences such as depression and poor health. Research suggests that many bullying incidents are not reported to school officials, which ham-pers educators’ ability to define the scope or frequency of bullying behavior in their schools or districts, the first step in addressing the problem. Further, when bullying is under-reported, administrators are likely to receive an incomplete picture of bullying behaviors in their schools and of the conditions and set-tings in which bullying occurs.

This study used nationally representative data from the 2007 National Crime Victimization

Survey School Crime Supplement, a biennial survey of children ages 12–18 who attended school in the prior academic year, to examine which of 51 characteristics of bullying victim-ization, bullying victims, and bullying victims’ schools are associated with increased report-ing of bullying to a teacher or other adult at the school. The survey data show that 36 percent of bullying victims reported their vic-timization to a teacher or other adult at their school and that 64 percent of students did not.

Eleven characteristics were found to have a statistically significant association with re-porting of bullying victimization, specifically:

• Eight characteristics of bullying victim-ization were statistically associated with increased reporting: bullying involving injury, physical threats, destruction of property, actual physical contact (pushing, shoving, and the like), greater frequency, multiple types, more than one location, and at least one occurrence on a school bus.

Seven characteristics did not appear to be associated with reporting: bullying that involved making fun of the victim or call-ing the victim names, excluding the vic-tim, spreading rumors about the victim,

ii Summary

and forcing the victim to do things he or she did not want to do, and bullying that occurred in the school building, on school grounds, or somewhere else.

• Three characteristics of bullying victims were found to have statistically significant relationships with reporting. Grade level was significantly and negatively associated with reporting, and being involved in a fight during the school year and being afraid of attack and avoiding certain school areas or activities were significantly and positively associated with reporting.

Victim characteristics that did not appear to be associated with reporting included gender, race/ethnicity, household region, and academic performance.

• No characteristic of bullying victims’ schools—including general characteristics, school culture, and school security and safety—was found to have a statistically significant association with reporting.

The results should be interpreted as exploratory associations between the reporting of bullying and various student and school characteristics and not as confirmations of causal relationships.

Regional Education Laboratory (REL) Northeast and Islands conducted this study of the conditions under which bullying victimization is reported in response to the concerns about bullying expressed by Parent Information and Resource Centers and other stakeholders in the REL Northeast and Islands Region and elsewhere.

Further research could be undertaken to understand why bullying is or is not reported and to learn more about the aftermath of reporting, including school responses to reports and whether victims who report bullying suffer reprisals. Such projects could require entirely new data collection efforts or the addition of items to existing student surveys such as the National Crime Victimization Survey School Crime Supplement.

August 2010

Table of conTenTs iii

Table of conTenTs

Why this study? 1Regional relevance 2National relevance 3Research questions 4

What the study found 6Characteristics of bullying victimization 6Characteristics of bullying victims 8Characteristics of bullying victims’ schools 10Summary of findings 15

Directions for future research 15

Appendix A Previous research on bullying 17

Appendix B Data source and methodology 19

Notes 34

References 35

Box

1 Data source and methodology 4

Figures

B1 Point-biserial correlation between the student fear of attack and avoidance of school areas and activities scale and reporting 30

B2 Point-biserial correlation between the school crime and drug problem scale and reporting 30

Tables

1 Relationship between bullying characteristics and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year 7

2 Relationship between sociodemographic characteristics and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year 9

3 Relationship between student school-related experience and perception and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year 11

4 Relationship between general school characteristics and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year 12

5 Relationship between school culture characteristics and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year 13

6 Relationship between school safety and security characteristics and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year 14

7 Characteristics of bullying victimization and bullying victims that were statistically significant in analyses 15

iv Table of conTenTS

B1 Bullying victimization items from the 2007 School Crime Supplement to the National Crime Victimization Survey used, created, or recoded 20

B2 Bullying victim items from the 2007 National Crime Victimization Survey and the School Crime Supplement used, created, or recoded 21

B3 Schools of bullying victim items from the 2007 School Crime Supplement to the National Crime Victimization Survey used, created, or recoded 22

B4 Unused items from the 2007 School Crime Supplement to the National Crime Victimization Survey 23

B6 Relationship between bullying characteristics and reporting of bullying, as indicated by students ages 12–18

B7 Kuder-Richardson coefficient of reliability for the student fear of attack and avoidance of school areas and

B5 Survey items with less than 95 percent response rate 26

during the 2007 school year, without proxy interviews 27

activities scale 30

B8 Kuder-Richardson coefficient of reliability for the school crime and drug problem scale 31

1 Why ThiS STudy?

This study tested 51 characteristics of bullying victimization, bullying victims, and bullying victims’ schools to determine which were associated with reporting to school officials. It found that 11 characteristics in two categories—bullying victimization and bullying victims— showed a statistically significant association with reporting. The study also notes the high percentage (64 percent) of respondents who experienced bullying but did not report it.

Why ThIs sTudy?

A student is bullied when he or she is “exposed, repeatedly and over time, to negative actions on the part of one or more other students” (Olweus 1993, p. 9). Bullying appears to be common among U.S. students and has been associated with short- and long-term negative consequences such as depression and poor health (Rigby 2003). Broad public concerns about the problems that appear to be associated with bullying have led school officials and others to attempt to mitigate such activity in their institutions.

Prior research suggests that many bullying incidents go unreported to school officials (see appendix A for a summary of previous research on bullying). Underreporting inevitably hampers educators’ ability to define the scope or frequency of bullying behavior in their schools or districts, the first step in addressing the problem. Further, when bullying is under-reported, administrators are likely to receive an incomplete picture of bullying behaviors in their schools and of the conditions and settings in which bullying occurs.

Learning more about reporting could assist educators in their decisionmaking. For example, data that indicate a large percentage of unreported bullying could lead educators to implement programs that facilitate victim and bystander reporting. Understanding more about the characteristics associated with the reporting of bullying victimization could inform schools of whether further interventions, such as education about bullying, are needed for students and staff. For example, some students may not report “indirect bullying”—such as being excluded or having rumors spread about them (DeVoe and Kaffenberger 2005)—because they do not view it as bullying or because they do not believe school staff would view it as such (Unnever and Cornell 2004). This study is a necessary step toward understanding more about the reporting of bullying to school officials.

2 characTeriSTicS of bullying, bullying vicTimS, and SchoolS aSSociaTed WiTh reporTing of bullying

Regional relevance

The implication of bullying as a factor in the suicides of students in the Northeast and Islands Region has drawn further attention to the problem of bullying in schools (see, for example, Associated Press 2009; Halligan 2005; King and Hendricks 2010; Marshall 2010; Vaznis 2009).

According to estimates from the states participating in the Centers for Disease Control and Prevention’s Youth Risk Behavior Survey in 2007, bullying affects a substantial share of the region’s students—from 17 percent in Vermont to 22 percent in Massachusetts to 29 percent in Connecticut. In addition, a recent school district survey in Newburyport, Massachusetts, found that 9–24 percent of students in grades 5 and 6 were victims of “frequent and persistent bullying” (Hendricks 2008). A 2007 statewide survey of elementary through high school girls in Vermont found that 17–30 percent self-reported victimization by bullies who used the Internet, cell phones, or other electronic means to inflict pain or embarrassment, a phenomenon known as cyber-bullying (Larkin 2007).

Many parent-based and other youth advocacy groups have been outspoken about the need to address bullying, and state legislatures have also taken considerable action in recent years. A majority of states have passed anti-bullying laws (Associated Press 2009), most of which mandate that schools or districts develop conduct codes that specifi

cally prohibit bullying, implement Within the northeast

and Islands Region,

antibullying laws have

been passed in new

hampshire (2000),

connecticut (2002),

Rhode Island (2003),

Vermont (2004), Maine

(2006), Puerto Rico (2008),

and Massachusetts

(2010) and are under

consideration in new york

strategies for dealing with bullying, and report all such incidents to the state education agency. Within the Northeast and Islands Region, such laws have been passed in New Hampshire (2000), Connecticut (2002), Rhode Island (2003), Ver-mont (2004), Maine (2006), Puerto Rico (2008), and Massachusetts (2010) and are under consideration in New York (Vaznis 2009).1 For example, in Vermont in 2004 the governor signed into law “An Act

Relating to Bullying Prevention Policies,” which requires all public and independent schools to have written rules for students prohibiting bullying behaviors, create clear policies for handling such incidents, and report all bullying events to the Vermont Department of Education. State education agencies have also taken action against bullying, including developing guides to assist schools and districts in dealing with it (see, for example, Maine Governor’s Children’s Cabinet 2006).

Given the widespread nature of bullying, several regional stakeholders have expressed interest in conducting research on the issue to Regional Educational Laboratory (REL) Northeast and Islands. Among the most vocal have been the parent information and resource centers, which were funded by the U.S. Department of Education beginning in 1995 to provide parents, schools, and organizations working with families with training, information, and technical assistance to understand how children develop and what children need to succeed in school. The Parent Advocacy Coalition for Education Rights Center, which serves as the national parent information and resource center, provides extensive resources on bullying (see www.pacer.org/bullying/). In addition, the parent information and resource center covering the Northeast and Islands Region has made bullying a priority issue through its relationship with the New Jersey Bar Foundation’s bullying prevention project and the New Jersey Coalition for Bullying Awareness and Prevention.2

This focus on bullying in the region’s public schools has also led to concern among educators and others that many bullying incidents are not reported to school officials. Highlighting this issue, an assistant principal at a Massachusetts high school was quoted in a recent article on bullying, as stating:

The problem for schools has always been that kids don’t report it. Students are afraid to report it because they’re afraid to escalate the problem. . . . Many times, it reaches a point, as it has recently, where the issue doesn’t come to light until it has gone too far (King and Hendricks 2010).

Why ThiS STudy? 3

Because most bullying occurs away from school officials, they depend on victim, bystander, and parent reports for incidents to come to their attention (Kazdin and Rotella 2009). The Massachusetts report, Direct from the Field: A Guide to Bullying Prevention, also underscores the concern about reporting:

The majority of bullying incidents happen outside of the eyes and ears of school personnel—on buses, on sidewalks on the way home, at sporting events, and in bathrooms and locker rooms. Complicity among young people not to share knowledge of incidents of bullying with adults is common, often due to fear of retaliation (Parker-Roerdon, Rudewick, and Gorton 2007, p. 6).

The regional parent information and resource center and United We Stand, a parent advocacy group for disabled students, also expressed considerable interest in the issue. The director of the regional center stated, “This is a very important issue for us. . . . Understand[ing] the most effective ways to encourage students to report bullying and harassment rather than seeing it as ‘tattletaling’ is critical.”3 The executive director of United We Stand, who is also a member of the REL Northeast and Islands Governing Board, encouraged the proposed project and stated that the findings would be of interest to her stakeholders.

National relevance

Maintaining safe schools is also a priority of the U.S. government, as underscored by federal legislation. For example, Title IV of the No Child Left Behind Act specifies funding for state education agencies to support school safety in the country’s schools. The law (20 USC 7131, Section 4121) reads:

(a) PROGRAM AUTHORIZED.—From funds made available to carry out this subpart under section 4003(2), the Secretary, in consultation with the Secretary of Health and Human Services, the Director of the Office of National Drug Control Policy, and

the Attorney General, shall carry out pro-grams to prevent the illegal use of drugs and violence among, and promote safety and discipline for, students. [emphasis added]

In addition, President Barack Obama, in his 2009 “Back to School” speech, stated:

Maybe you’ll decide to stand up for kids who are being teased or bullied because of who they are or how they look, because you believe, like I do, that all kids deserve a safe environment to study and learn (The White House 2009).

The National Safe Schools Partnership, a coalition of nearly 30 education, health, and other organizations promoting federal legislation to advance safe schools, has described bullying and harassment as a “prevalent and profound” problem (2007, p. 1).4

National estimates of bullying vary, but Dinkes, Kemp, and Baum (2009) found that some 32 percent of school children ages 12–18 self-reported having been bullied during the previous school year.

And although the relationship between bullying and school performance is complex (see, for example, Pepler and Craig 2008), the widespread nature of bullying counters emphasis on school safety and discipline by the U.S. Department of Education and the No Child Left Behind Act and may be a roadblock to some students’ adequate academic achievement (Srabstein and Piazza 2008; Glew et al. 2005; Juvonen, Graham, and Schuster 2003). A wide range of stakeholders outside education have also taken up the issue, including the American Academy of Pediatrics (Klass 2009), the U.S. Department of Justice’s Office of Community Oriented Policing Services (Sampson 2004), and the Canadian Psychological Association (2009).

although the

relationship between

bullying and school

performance is complex,

the widespread nature

of bullying counters

emphasis on school

safety and discipline by

the u.s. department of

education and the no

child left behind act and

may be a roadblock to

some students’ adequate

academic achievement


The national attention on bullying includes concern about whether incidents get reported to school officials. Many students do not report that they have been bullied (Unnever and Cornell 2004), and officials are unable to take action to address individual incidents to protect victims (Pepler and Craig 2008). Moreover, educators are often unaware of the scope of the bullying problem, hindering efforts to base policies and programs on sound data (Unnever and Cornell 2004). Concern about reporting is evidenced by the number of school districts that have moved to an anonymous hotline reporting system, hoping that such a mechanism would remove student fear of reprisal and encourage more reporting (Teicher 2006; Peterson 2009).

Research questions

Based on bullying victims’ reports of whether their victimization was reported to school officials and based on REL Northeast and Islands

stakeholders’ interest in understanding under-reporting of bullying, the following research questions were addressed:

• What characteristics of bullying victimization are associated with increased reporting of bullying to a teacher or other adult at the school?

• What characteristics of bullying victims are associated with increased reporting of bullying to a teacher or other adult at the school?

• What characteristics of bullying victims’ schools are associated with increased reporting of bullying to a teacher or other adult at the school?

Data sources and methodology are described briefly in box 1 and detailed in appendix B. The study was informed by a review of the current literature (see appendix A).

box 1

Data source and methodology

To respond to the research questions, secondary analysis of data from the U.S. Department of Justice, Bureau of Justice Statistics’ 2007 National Crime Victimization Survey School Crime Supplement was conducted. Hagan (1993, p. 215) defines secondary analysis as the “re-analysis of data that were previously gathered for other purposes.”

Data source. The National Crime Victimization Survey is a nationally representative survey administered annually by the U.S. Census Bureau on behalf of the Bureau of Justice Statistics to persons ages 12 and older in selected households across

the contiguous United States. Every other year, the survey includes the School Crime Supplement, which covers all students ages 12–18 who attended at least some school in the prior academic year. The 2007 survey invited 11,161 people ages 12–18 to participate; 6,503 of them completed the survey, and 5,621 met the screening criteria and thus comprise the data set used to conduct the secondary analysis.

Identifying reported and unreported bullying. Students were considered bullied if they responded affirmatively to having been bullied in one or more of the following ways: being made fun of; being the subject of rumors; being threatened with harm; being pushed, shoved, tripped, or spit

on; being made to do things they did not want to do; being intentionally excluded from activities; and having property intentionally destroyed.

To identify whether bullied students’ victimization was reported to a teacher or other adult at the school, students were asked, “Was a teacher or some other adult at the school notified about (this event/any of these events)?” The question does not indicate who reported the bullying victimization.

Handling survey nonresponse and complex survey sampling. Two issues with sample construction were taken into account. First, not all students eligible to respond to the survey participated, which could bias

(conTinued)

5

box 1 (conTinued)

Data source and methodology

results if those who responded differ substantively from those who did not. According to the U.S. Department of Justice (2007), nonparticipating students are more likely to come from racial/ethnic minority, urban, and lower income households, so the data are weighted accordingly (see appendix B). Second, the survey uses a stratified, multistage cluster sample design. To avoid biased estimates in the analysis, this complex sampling design required using sample weights, sampling units (clusters), and sampling strata to adjust for clustering and stratification to compute valid standard errors.

Selecting items for analysis. Of the 140 items in the National Crime Victimization Survey School Crime Supplement, 51 in three domains (bullying victimization, bullying victim, and bullying victims’ schools) were included in this study. The bullying victimization domain included 15 items, such as whether injury to the victim occurred, types of bullying involved (direct or indirect), bullying severity and frequency, and the locations in which it occurred. The bullying victim domain included 14 items, such as sociodemographic characteristics, grade level, household region, current academic performance, whether the student has an adult at school who cares about him or her, and a scale on how fearful the student is of being attacked. The bullying victims’ schools domain included 22 items, such as whether the school was public or private; perceptions of the school’s rules and how they are

enforced; perceptions of how teachers treat students; a scale of school crime and drug problems; and perceptions of school safety. Tables B1–B4 in appendix B provide a complete list of the items that were selected, recoded, or created for the analysis.

A few selected items from the household portion of the larger National Crime Victimization Survey, such as household income and region, were also included.

Conducting statistical analyses. Descriptive analysis was conducted to respond to the research questions, focusing on comparisons between reported and unreported bullying according to victim self-reports. Cross-tabulations (usually 2×2 tables) analyzed the presence or absence of a characteristic with reporting or nonreporting. Differences between reporters and nonreporters were tested using Pearson’s chi-square because the variables were categorical. Since chi-square analysis does not indicate the direction of the relationship between variables, correlations were calculated for statistically significant items to determine whether a variable was associated with an increase or decrease in reporting. Point-biserial (rpbi) correlations were used to indicate directionality for the two scales (student fear of being attacked and school crime and drug problems) and other continuous variables. For statistically significant dichotomous variables, tetrachoric correlations (rho) are reported to

Why ThiS STudy?

indicate directionality (Welkowitz, Ewen, and Cohen 1982).

The initial threshold to determine statistical significance was p = .05 (two-tailed). But because of the number of significance tests conducted, there is an increased likelihood of some results being statistically significant due to chance. A Bonferroni multiple comparison procedure was calculated to adjust for the number of significance tests (Bland and Altman 1995). These adjusted statistical significance levels were used to identify statistically significant associations.

Limitations. All the data analysis is descriptive and does not allow for causal interpretation. No conclusions about the effectiveness of school policies and strategies on the reporting of bullying can be reached. The National Crime Victimization Survey School Crime Supplement presents data on a wide range of school crime, safety, and discipline issues; it is not focused specifically on bullying. It contains only one item on whether the bullying experienced was reported to an adult, and that item is not linked to any specific bullying incident or time sequence. The survey also relies on respondents to self-determine their condition as a victim of bullying using their own interpretation and conceptions of bullying. Different respondents might not label similar situations as bullying. And some students may be reluctant to tell an interviewer about being bullied, so some victims may not be included in these analyses.


WhaT The sTudy found

Findings are organized into three sections (characteristics of bullying victimization, characteristics of bullying victims, and characteristics of bullying victims’ schools) to respond directly to the three research questions. The survey data show that 36 percent of bullying victims reported their victimization to a teacher or other adult at their school and that 64 percent of students did not.

Characteristics of bullying victimization

This section presents the analysis of the relationship between characteristics of bullying victimization and reporting. The bullying characteristics included in the analysis were:

• Whether the bullying caused injury to the victim.

• The type of bullying that occurred (threats, destroyed property, physical violence, victim being made fun of or called names, victim being excluded, victim having rumors spread about him or her, and victim being made to do things he or she did not want to do).

• How many types of bullying the victim experienced.

• The frequency of the bullying.

• The location where the bullying occurred (school building, outside school grounds, school bus, and somewhere else).

• The number of different locations where the bullying took place.

Eight characteristics showed a statistically significant relationship with reporting; seven did not. Table 1 provides data on overall reporting of bullying and on reporting across the six types of

bullying captured by the National Crime Victimization Survey School Crime Supplement. The categories are not mutually exclusive, as bullying victimization over the year may include several types. The reporting item is not linked to any specific incident or point in time, so these data represent reporting of bullying that involves, at the very least, that particular type.

The severity of victimization showed a statistically significant and positive association with reporting, meaning that the presence of this type of bullying was associated with increased reporting (tetrachoric rho = .33, p < .001). When bullying resulted in physical injury, 60.5 percent of bullying victims indicated that their victimization was reported. Bullying that involved a physical threat was reported 55.3 percent of the time, bullying that involved destroyed property was reported 51.3 percent of the time, and bullying that involved being physically touched (pushed, shoved, or tripped) was reported 46.0 percent of the time. Such direct types of bullying (DeVoe and Kaffenberger 2005) also showed a statistically significant and positive association with reporting, meaning that the presence of each of these types of bullying was associated with increased reporting (tetrachoric rho = .35 for threats, .23 for destroyed property, and .25 for being pushed, shoved, or tripped).

The number of types of bullying experienced and reporting showed a statistically significant and positive relationship, meaning that a higher number of types of bullying experienced was associated with increased reporting (rpbi = .20, p < .001). Reporting rates ranged from 25.7 percent for students who indicated that they were victims of one type of bullying to 59.4 percent for students who indicated that they were victims of six types of bullying.

The relationship between bullying frequency and reporting was also statistically significant and positive, meaning that increased frequency of bullying was associated with increased reporting (rpbi =.19, p < .001). For example, 44.9 percent of victims who were bullied once or twice a week said the bullying was reported to school officials, and

eight characteristics of

bullying victimization

showed a statistically

significant relationship

with reporting;

seven did not

Table 1

Relationship between bullying characteristics and reporting of bullyingduring the 2007 school year

, as indicated by students ages 12–18

Students Students whose whose

bullying was bullying was number of reported not reported

characteristic observations (percent) (percent) Standard

error -chi square

value

chance probability

of result

victim of bullying

victim of bullying 1,778 35.8 64.2 — — —

physical injury to victim

yes 119 60.5 39.5 5.19 32.8* <.001

no 1,657 34.2 65.8 1.26

Type of bullying

Threatened 323 55.3 44.7 2.93 64.93* <.001

destroyed property 231 51.3 48.7 3.54 27.97* <.001

pushed, shoved, tripped, and the like 626 46.0 54.0 2.09 41.80* <.001

made fun of, called names 1,180 38.2 61.8 1.42 7.43 .008

excluded 301 37.7 62.3 2.93 .514 .522

Spread rumors 1,010 37.5 62.5 1.55 2.22 .173

victim made to do things he or she did not want to do 232 37.3 62.7 3.19 .22 .641

number of types of bullying experienced

one 714 25.7 74.3 1.73

68.42* <.001

Two 487 37.9 62.1 2.32

Three 286 44.6 55.4 3.17

four 163 45.5 54.5 3.71

five 80 53.8 46.2 5.33

Six 31 59.4 40.6 8.61

Seven 17 50.3 49.7 12.16

frequency of bullying during academic year

once or twice this school year 1,060 32.6 67.4 1.66

20.58* <.001 once or twice a month 352 39.7 60.3 2.33

once or twice a week 170 44.9 55.1 3.47

almost every day 112 48.5 51.5 4.70

location where bullying occurred

School building 1,401 36.7 63.3 1.37 1.74 .187

outside on school grounds 407 38.6 61.4 2.52 1.63 .216

School bus 146 47.8 52.2 4.19 9.48* .002

Somewhere else 70 26.7 73.3 5.31 2.67 .098

number of different locations bullying occurred

one 1,558 34.8 65.3 1.25 11.88* .001

Two or more 202 47.1 52.9 3.73

— is not applicable.

* Difference between characteristic of bullying victimization and reporting is statistically significant, p < .0033.

Source: Authors’ analysis based on U.S. Department of Justice 2007.

7 WhaT The STudy found


one sociodemographic

characteristic of

bullying victims


significant correlation

with reporting; four

did not. Two school-

related experience

and perception

characteristics showed

a statistically significant

relationship with

reporting; seven did not

48.5 percent of victims who were bullied almost every day said the bullying was reported to school officials.

The relationship between location and reporting was also statistically significant. Bullying victimization that included at least one occur-rence on a school bus (tetrachoric rho = .18, p =.002) or occurred in multiple locations (rpbi = .08, p =.001) was associated with increased reporting.

Bullying that involved the victim being made fun of or called names, the victim being excluded, the victim having rumors spread about him or her, or the victim being made to do things that he or she did not want to do were not associated with increased reporting.

Characteristics of bullying victims

This section presents the analysis of the relationship between characteristics of bullying victims and reporting. Two types of victim characteristics were included: sociodemographic characteristics and student school-related experiences and perceptions.

Sociodemographic characteristics. Victim sociodemographic characteristics included in the analysis were:

• Gender.

• Race/ethnicity.

• Grade level.

• Household region.

• Household income.

One characteristic showed a statistically significant correlation with reporting; four did not.

Reporting by grade level ranges from 52.9 percent for students in grade 6 to 27.0 percent for students in grade 12. The relationship between grade level and reporting of bullying is statistically significant and negative, meaning that higher grade levels are associated with less reporting (rpbi = –.18, p < .001).

Male and female bullying victims did not differ in the prevalence of reporting (table 2)—that is, the percentage of girls who indicated that their bullying victimization was reported to a teacher or other adult at the school did not statistically differ from the percentage of boys who indicated that their victimization was reported. Moreover, reporting did not statistically differ across racial/ ethnic groups.

The region of the country (as defined by the Census Bureau) where the student’s household is located did not affect reporting. Students from households in the Northeast and students from all other regions indicated that similar percentages of bullying victimization were reported to school officials (35.7 percent compared with 36.0 percent).

No statistically significant association was found between household income and reporting. This is one of the few variables in the data set used in the analyses that had an item response rate lower than 95 percent (see table B1 in appendix B), so the results for household income should be interpreted with caution.

School-related experiences and perceptions. Student school-related experiences and perceptions included in the analysis were:

• Victim’s academic performance.

• Whether the victim skipped classes during the academic year.

• Whether the victim has an adult at school who cares about him or her.

• Whether the victim’s school has an adult who helps him or her with problems.

Table 2

Relationship between sociodemographic characteristics and reporting ages 12–18 during the 2007 school year

of bullying, as indicated by students




error -chi square

value

chance probability

of result

gender

male 874 34.5 65.5 1.66 1.57 .213

female 904 37.3 62.7 1.66

race/ethnicity

non-hispanic White 1,193 33.1 66.9 1.35

13.49 .015 non-hispanic black 221 44.4 55.7 3.34

non-hispanic other 101 37.2 62.8 5.63

hispanic 263 39.4 60.6 3.56

current grade

6 221 52.9 47.1 3.32

66.18* <.001

7 311 45.9 54.1 2.92

8 322 36.5 63.6 2.93

9 291 27.3 72.7 2.86

10 246 27.9 72.1 3.05

11 238 30.6 69.4 3.49

12 149 27.0 73.0 3.86

region where student household located

northeast 267 35.7 64.3 3.41 .009 .939

all other 1,511 36.0 64.0 1.28

household income

less than $7,500 47 45.3 54.7 7.26

18.08 .005

$7,500–$15,000 53 54.5 45.5 6.28

$15,001–$25,000 124 39.3 60.7 4.39

$25,001–$35,000 152 32.4 67.5 4.14

$35,001–$50,000 258 38.1 61.9 3.21

$50,001 or more 792 31.7 68.3 1.72

* Difference between the characteristic of bullying victims and reporting is statistically significant, p < .0033.



• Whether the victim has a friend at school to talk to.

• Whether the victim has a friend at school who helps him or her with problems.

• Whether the victim was involved in a fight during the school year.

• Whether the victim brought a weapon to school.

• How much the victim fears attack and avoids school areas or activities.

Two characteristics showed a statistically significant relationship with reporting; seven did not.


It is not possible to determine whether fights that a victim was involved in were related to bullying incidents, based on the National Crime Victimization Survey Student Crime Supplement data set. But student-reported fighting behavior was significantly and positively associated with the reporting of bullying behavior, meaning that having been involved in a fight was associated with increased reporting (tetrachoric rho = .30, p < .001). Specifically, 54.1 percent of students who responded that they were involved in fighting behavior during the past academic year indicated that their bullying victimization was reported, compared with 32.8 percent of students who responded that they had no involvement in fighting during the past academic year. (table 3).

To measure whether the victim fears attack or avoids school areas or activities, a student fear of attack and avoidance of school areas or activities scale was created, comprising 14 items. Students were asked three items about their fear at school, their fear on the way to or from school, and their fear about being attacked or harmed outside of school. For these three items, students indicated whether they were never afraid, almost never afraid, sometimes afraid, or afraid most of the time. “Never afraid” and “almost never afraid” responses counted for 0 points on the scale, and “sometimes afraid” and “afraid most of the time” responses counted for 1 point on the scale. Students were then asked 11 items about whether they avoided school, certain activities at school, or certain locations in the school because of their fear of attack. Each location or activity that a student avoided because of fear of attack counted for 1 point on the scale. The relationship between the scale and reporting was statistically significant and positive, meaning that a higher score on the scale was associated with increased reporting (rpbi = .12).

Academic performance was measured based on average course grades. Higher grades (A’s, B’s, and C’s) were combined and compared with lower grades (D’s and F’s). There was no relationship between students’ self-reported academic grades and reporting.

Because of the research design of the Student Crime Supplement, it is not possible to determine whether skipping classes was a direct result of having been bullied, but no statistical relationship was found between student responses to this item and reporting.

The National Crime Victimization Survey Student Crime Supplement asks students to indicate how strongly they agree or disagree with four statements: they have an adult at the school who cares about them, their school has an adult that helps them with problems, they have a friend at school they can talk to, and they have a friend at the school who helps them with their problems. Students were asked to indicate whether they strongly agree, agree, disagree, or strongly disagree with each statement. A substantial majority of students agreed or strongly agreed with all four statements. None of the items met the threshold for statistical significance.

Students were also asked whether they had ever brought a gun, knife, or other weapon to school. The association between weapon carrying and reporting was not statistically significant.

Characteristics of bullying victims’ schools

This section presents the analysis of the relationship between characteristics of bullying victims’ schools and reporting. Three types of variables were included: general school characteristics, school culture characteristics, and school safety and security measures

General school characteristics. The general school characteristics included in the analysis were:

• Whether school is public or private.

• Whether the school is church-related.

Most students in the sample attended public schools. Some 36.4 percent of bullying victims attending public schools indicated that their victimization was reported to school officials, 30.3 percent of students attending private schools indicated that

none of the

characteristics of

bullying victims’ schools


significant relationship

with reporting

Table 3

Relationship between student school-related experience and perception and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year


bullying was bullying was chance number of reported not reported Standard -chi square probability

experience or perception observations (percent) (percent) error value of result

academic performance

mostly a’s, b’s, and c’s 1,652 35.6 64.5 1.23 1.05 .326

mostly d’s and f’s 105 40.5 59.5 1.23

Skipped classes during academic year

yes 164 27.4 72.6 3.93 5.60 .038

no 1,588 36.8 63.2 1.30

adult at school who cares about me

Strongly agree or agree 1,599 36.5 64.5 1.26 2.35 .120

Strongly disagree or disagree 175 30.6 69.4 3.52

School has adult who helps with problems



have friend at school to talk to



have friend at school who helps with problems



involved in a fight

yes 258 54.1 45.9 3.56 43.1* <.001

no 1,519 32.8 67.2 1.29

brought a weapon to school

no 1,709 36.2 63.8 1.22 .392 .531

yes 65 32.3 67.7 5.87

Student fear of attack and avoidance of school areas or activities scale score

0–2 21 43.4 56.6 11.08

29.5* <.001 3–5 1,636 34.3 65.7 1.233

6–8 95 53.7 46.3 9.99

9 or higher 20 36.0 64.0 1.2.1

* Difference between characteristic of bullying victims and reporting is statistically significant, p < .0033.

Source: Authors’ anyalsis based on U.S. Department of Justice 2007.


their victimization was reported, and 29.2 percent of students attending church-related schools indicated that their victimization was reported (table 4). The relationship between type of school and reporting was not statistically significant.

School culture characteristics. The school culture characteristics measured how much students agreed or disagreed (strongly agree, agree, disagree, or strongly disagree) with eight statements about their school:

Table 4

Relationship between general school characteristics and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year



characteristic observations (percent) (percent) error value of result

School type

public 1,639 36.4 63.6 1.25 2.10 .161

private 139 30.3 69.7 4.03

church-related school

yes 105 29.2 70.8 4.74 2.27 .167

no 1,673 36.4 63.6 1.26



• Everyone knows school rules.

• Students receive the same punishment for breaking the same rules.

• Students know the punishments.

• School rules are fair.

• School rules are strictly enforced.

• Teachers care about students.

• Teachers treat students with respect.

• Teachers make students feel bad.

The school culture characteristics also measured students’ opinions of how often (never, almost never, sometimes, or most of the time) two actions related to classroom misbehavior occur:

• Student is distracted by students misbehaving in class.

• Teachers punish students for misbehaving in class.

“Strongly agree” and “agree” responses were collapsed into one category, “disagree” and “strongly

disagree” responses were collapsed into one category, “never” and “almost never” responses were collapsed into one category, and “sometimes” and “most of the time” responses were collapsed into one category.

There was no statistically significant association between whether bullying victims agree or disagree with any of the statements and reporting, nor was there one between students’ opinions of how often they were distracted by other students misbehaving in the classroom and reporting or between students’ opinions of how often teachers punished students for misbehaving in class and reporting (table 5).

School safety and security measures. The school safety and security characteristics included in the analysis were:

• Whether the school has security guards.

• Whether the school has staff or adults monitoring the hallway.

• Whether the school has metal detectors.

• Whether the school has locked doors.

• Whether the school has a visitor sign-in policy.

• Whether the school conducts locker checks.

Table 5

Relationship between school culture characteristics and reporting of bullying, as indicated by students ages 12–18 during the 2007 school year



characteristic observations (percent) (percent) error value of result

everyone knows school rules


disagree or strongly disagree 242 40.6 59.4 3.63

Students receive the same punishment for breaking the same rules

Strongly agree or agree 1,314 35.9 64.1 1.39 .004 .957


Students know the punishments



School rules are fair



School rules are strictly enforced



Teachers care about students



Teachers treat students with respect



Teachers make students feel bad

Strongly agree or agree 477 36.5 63.5 2.09 .090 .743

disagree or strongly disagree 1,297 35.7 64.3 1.33

how often distracted by students misbehaving in class

never or almost never 448 30.9 69.1 2.65 6.88 .031

Sometimes or most of the time 1,327 37.7 62.3 1.39

how often teachers punish students for misbehaving in class

never or almost never 491 31.0 69.0 2.47 7.27 .024

Sometimes or most of the time 1,281 37.8 62.2 1.44



• Whether the school has safety badges. Analysis was conducted for students that responded “yes” or “no” to each item. Although

• Whether the school has security cameras. “don’t know” was a valid response option, it was excluded. Several items thus have response rates

• Whether the school has a student code of conduct. below 95 percent (see table B5 in appendix B).

Table 6

Relationship between school safety and security charactestudents ages 12–18 during the 2007 school year

ristics and reporting of bullying, as indicated by

Students whose

bullying was number of reported

characteristic observations (percent)

Students whose

bullying was not reported

(percent) Standard

error -chi square

value

chance probability

of result

School has security guards

yes 1,174 34.3 65.7 1.46 3.86 .056

no 586 39.1 60.9 2.14

School has staff or adults monitoring the hallway

yes 1,571 35.9 64.1 1.31 .046 .848

no 196 36.7 63.4 3.82

School has metal detectors

yes 133 35.1 65.0 3.74 .060 .797

no 1,568 36.1 63.9 1.30

School has locked doors

yes 1,058 36.4 63.6 1.61 .027 .866

no 628 36.8 63.2 1.23

School has visitor sign-in policy

yes 1,679 36.1 63.9 1.27 1.30 .314

no 69 42.9 57.1 1.24

School has locker checks

yes 1,006 34.4 65.6 1.6 3.15 .075

no 665 38.7 61.3 1.90

School has safety badges

yes 378 38.5 61.5 2.50 1.48 .242

no 1,397 35.2 64.9 1.39

School has security cameras

yes 1,220 35.0 65.0 1.52 1.00 .327

no 377 37.9 62.1 1.36

School has a student code of conduct

yes 1,727 35.7 64.3 1.26 .798 .379

no 37 42.9 57.1 8.14

School crime and drug problem scale score

0 249 43.3 56.7 3.47

7.17 .293

1 240 38.2 61.8 3.52

2 128 40.7 59.3 4.25

3 104 37.1 62.9 5.59

4 75 35.8 64.2 5.80

5 or higher 362 33.1 66.9 2.62



direcTionS for fuTure reSearch 15

The school safety and security characteristics also include a measure of students’ perception of crime and drug problems at their school. The school crime and drug problem scale comprises 13 items: whether the student knew other students who brought a loaded gun to school, whether he or she had seen another student with a loaded gun at school, whether he or she could have gotten a loaded gun at school, whether gangs were at the school, whether gangs were involved in selling drugs at school, whether gangs at school were involved in fights and violence, whether he or she had seen hate-related words and symbols at school, whether he or she was offered drugs or alcohol during the academic year, whether he or she knew other students on drugs or alcohol, whether it was possible to get alcohol at school, whether it was possible to get marijuana at school, whether it was possible to get prescription drugs at school, and whether it was possible to get crack, cocaine or other drugs at school. Each “yes” response counted for 1 point on the scale.

None of the school safety and security measures showed a statistically significant relationship with reporting (table 6).

Summary of findings

Table 7 summarizes the findings, presenting the 11 of 51 characteristics found to have a

statistically significant (meeting the conservative statistical significance levels set by the Bonferroni procedure to counter the problem of conducting multiple statistical significance tests) association with the reporting of bullying, including eight bullying victimization characteristics and three student victim characteristics. None of the characteristics of bullying victims’ schools were found to have a statistically significant association with reporting.

future projects could

examine the aftermath

of reporting for bullying

victims or explore why

such a high percentage

of bullying victimization

is not reported

dIRecTIons foR fuTuRe ReseaRch

This study focused on the characteristics associated with the reporting of bullying to school officials. The survey data show that 35.8 percent of bullying victims indicated that their victimization was reported to a teacher or other adult at their school and that 64.2 percent of students did not. Future projects could examine the aftermath of reporting for bullying victims. Such a project could also explore why such a high percentage of bullying victimization is not reported (for example, fear of retaliation by bullies or belief that the school cannot help).

Table 7

characteristics of bullying victimization and bullying victims that were statistically significant in analyses

characteristics of bullying victimization characteristics of bullying victims

physical injury to victim (+) current grade (–) Threatened (+) involved in a fight (+) destroyed property (+) Student fear of attack and avoidance of school areas and pushed, shoved, tripped, and the like (+) activities (+) number of types of bullying experienced (+) frequency of bullying during academic year (+) bullying occurred on school bus (+) bullying occurred at more than one location (+)

+ indicates a positive relationship, meaning that the variable (or an increase in the variable’s value, for discrete variables) leads to an increase in reporting.

– indicates a negative relationship, meaning that the variable (or an increase in the variable’s value, for discrete variables) leads to a decrease in reporting.

Note: None of the characteristics of bullying victims’ schools were found to have a statistically significant association with reporting.



Items could also be added to the National Crime Victimization Survey School Crime Supplement or other national surveys (such as the Youth Risk Behavior Surveillance System) to probe deeper into what happens following the reporting of bullying to school officials. For example, what did the school do in response? Did the victim suffer reprisals? Items that distinguish who actually reported the bullying to a school official (the victim, bystander, or parent), that identify whether students told their parents and what the parental response was, and that list the reasons bullying victims have for reporting or

not reporting to school officials would also be useful, as would items that ask students whether they witnessed bullying and what they did about it. These data could improve the research evidence relevant to bystander behavior and school bullying.

The 2009 National Crime Victimization Survey School Crime Supplement data should be available in 2011. Repeating the analysis of bullying, victim, and school characteristics with the 2009 data would provide information about how the findings replicate or change over time.

17

aPPendIx a PReVIous ReseaRch on bullyIng

Agreement on how to define bullying is elusive (Griffin and Gross 2004). Olweus (1993, p .9) defines bullying as being “exposed, repeatedly and over time, to negative actions on the part of one or more other students,” a definition adopted by DeVoe and Kaffenberger in their National Center for Education Statistics report (2005, p. v). In another National Center for Education Statistics document, Dinkes, Kemp, and Baum (2009, p. 40) state that bullying includes “being made fun of; being the subject of rumors; being threatened with harm; being pushed, shoved, tripped, or spit on; being pressured into doing things did not want to do; [being] excluded; and having property destroyed on purpose.”

National estimates of bullying vary. Some 16 percent of students participating in a National Institute of Child Health and Development survey in 1998 stated that they had been bullied in their current school term (National Institutes of Health 2001). The 2008 Indicators of School Crime and Safety Report (Dinkes, Kemp, and Baum 2009), however, estimates that 32 percent of children nationwide were victims of bullying in 2007, and that 24 percent of public schools reported that student bullying was a daily or weekly problem during the 2005/06 school year. A 2003 national survey of parents indicated that 35 percent were worried about their child being bullied and 24 percent reported that their own child bullies or is cruel to other children (Sidorowicz, Hair, and Milot 2009). In a Kaiser Family Foundation (2001) survey of more than 800 students, bullying, teasing, and “put downs” were rated together as the number one problem in school (Boorstein 2004; CNN 2001).

Research suggests a number of potential negative consequences of bullying. Rigby’s (2003) review of this work summarizes the harms by type of research. For example, in cross-sectional surveys, victims of bullying report higher levels of depression and poor health than do nonvictims

appendix a. previouS reSearch on bullying

(Srabstein and Piazza 2008; Fekkes, Pijpers, and Verloove-Vanhorick 2004). Retrospective interview and questionnaire studies suggest that bullying contributes to victims’ difficulties with physical and psychological health, even into adulthood (Fosse and Holen 2002). These effects are more strongly substantiated in longitudinal studies that have reported bullying as a significant factor in students’ negative health and well-being and suggest that the consequences of bullying can be long term (Sourander et al. 2000). Longitudinal studies also identify that being a bully is a predictor of later involvement in antisocial and criminal behavior (Sourander et al. 2007). As mentioned, the relationship between bullying and academic achievement is complex, but some studies report negative academic performance for both victims and bullies (Olweus 1993; Farrington and Ttofi 2009).

Some studies have also suggested a link between bullying victimization and suicide and homicide. One study found that boys and girls who are bullied are four to eight times more likely to kill themselves than are nonvictims (Fox et al. 2003). Moreover, the Secret Service documented bullying victimization in the backgrounds of approximately two-thirds of attempted or completed school shooting attackers (Vossekull et al. 2002). Lawsuits have been brought against schools and districts for not doing enough to keep bullied children safe (Dawson 2006; Martindale 2009).

Though once considered by many adults as a normal adolescent rite of passage (Garbarino and DeLara 2003), the potential short- and long-term consequences of bullying have raised concern among administrators, teachers, parents, pediatricians, police, and others (National Crime Prevention Council 2008; National Safe Schools Partnership 2007). Such concern includes the aforementioned legislation in at least 44 states mandating that schools track incidents of bullying and take measures to address it (Associated Press 2009).

One major problem for concerned adults, however, is that bullying often goes unreported to teachers


or other school officials. Only 35.8 percent of bullied students in the 2007 National Crime Victimization Survey School Crime Supplement indicated that their bullying victimization was reported to school officials (Dinkes, Kemp, and Baum 2009). In a survey of more than 2,000 Dutch elementary school students, 16 percent reported having been bullied during a six month period; 53 percent of these victims reported the bullying to their teacher and 67 percent to parents (Fekkes, Pijpers, and Verloove-Vanhorick 2005). A survey administered by the Oklahoma Department of Health reported that 67 percent of students in grade 3, 47 percent of students in grade 5, and 20 percent of students in grade 7 who were bullied told an adult at the school (Middleton 2008).

Reporting is an important precursor to school response to bullying. Kazdin and Rotella (2009) note that teachers observe only the most flagrant and frequent bullying, and estimate that teacher observation occurs in only about 4 percent of incidents. Along with victim reluctance, bystanders who witness bullying also tend not to report it, even though 85 percent of incidents occur in front of others, usually peers (Kazdin and Rotella 2009).

Underreporting of bullying makes it difficult for school officials, parents, and other concerned adults to learn about and deal effectively with victimization (Education Development Center 2008).

Oliver and Candappa (2007) found that students are reluctant to tell adults about bullying and that this reluctance increases with age.

Little research on the reporting of bullying to school officials is available to guide stakeholders in the Northeast and Islands Region and elsewhere, particularly research conducted in the United States. REL Northeast and Islands researchers found one study that examined the differences between reported and nonreported incidents. Unnever and Cornell (2004) surveyed six middle schools (grades 6–8) in Roanoke, Virginia. Of the 2,437 students who participated, 898 (37 percent) were identified as bullying victims. Of the bullying victims, 25 percent did not report their victimization to anyone and 40 percent did not report it to an adult. Unnever and Cornell (2004) then analyzed which factors influenced victim reporting and found that victims who were bullied more frequently and by a larger number of bullies, who were female, who perceived that their school would not tolerate bullying, and who were from the lower grade levels were more likely to report. To better inform education decisionmakers in the region and elsewhere, further studies like this are needed. This REL Northeast and Islands project expands on the Roanoke study to empirically study differences between reported and nonreported bullying victimization, using a nationally representative data set.

19 appendix b. daTa Source and meThodology

aPPendIx b daTa souRce and MeThodology

This appendix provides more detail on the data source and methodology used for this study. To respond to the research questions, secondary analysis of data from the U.S. Department of Justice, Bureau of Justice Statistics’ 2007 National Crime Victimization Survey School Crime Supplement was conducted. Hagan (1993, p. 215) defines secondary analysis as the “re-analysis of data that were previously gathered for other purposes.”

Data source

The National Crime Victimization Survey is a nationally representative survey administered annually by the U.S. Census Bureau on behalf of the Bureau of Justice Statistics to persons ages 12 and older in selected households across the contiguous United States. The purpose of the survey is to get at the “hidden figure” of crime. Many crimes go unreported to the police, so relying on such reports to establish crime rates (as is done when using the Federal Bureau of Investigation’s summaries of “reports to the police,” known as the Uniform Crime Reports) provides a limited picture of criminal victimization (Hagan 1993).

Every other year the School Crime Supplement is added to the National Crime Victimization Survey on behalf of the U.S. Departments of Justice and Education. The supplement covers all students ages 12–18 who attended at least some school in the prior academic year. The 2007 survey invited 11,161 people ages 12–18 to participate; 6,503 of them completed the survey, and 5,621 met the screening criteria and thus comprise the data set used to conduct the secondary analysis. The purpose of the supplement is to provide a fuller picture of victimization beyond that captured by official reports to police of crimes at school. It asks approximately 140 items on a wide range of school behaviors and student perceptions, several of which deal specifically with bullying. Tables B1–B4 list the survey items used for the analysis in this study by category of characteristic. A full

copy of the survey instrument can be found in the supplement’s codebook (U.S. Department of Justice 2009).

The 2007 National Crime Victimization Survey is the best available source of data to examine reporting of bullying. Other possible data sources do not include enough information on bullying to adequately respond to the research questions. For example, each state in the Northeast and Islands Region participates in the Youth Risk Behavior Surveillance program overseen by the U.S. Centers for Disease Control and Prevention. But its core survey instrument did not include any items about bullying until 2009. States are free to add their own additional items, and three states in the region have added two items on bullying (“Have you been bullied in the past six months?” and “Have you bullied someone in the past six months?”). But no items about reporting bullying to school officials are included in the core or individual state Youth Risk Behavior Surveillance instruments.

Several states in the region have passed anti-bullying legislation that requires schools to collect information on bullying and transmit the data to the state department of education. Disciplinary files may also capture reports to school officials about bullying behavior. But even if these data files were accessible and contained reliable and comprehensive data on reporting of bullying, they would seriously underrepresent bullying, given that most bullying incidents are not reported to officials (at least 64 percent according to the 2007 National Crime Victimization Survey). These data files also provide no opportunity to contrast students reporting victimization with students who did not report their victimization to school officials.

Summary statistics on bullying using previous years’ National Crime Victimization Survey School Crime Supplement data are produced annually for the National Center for Education Statistics Indicators of School Crime and Safety publication (Dinkes, Cataldi, and Lin-Kelly 2008). In addition, general bullying statistics are made available using the “quick tables” function on the National


Table b1

bullying victimization items from the 2007 school crime supplement to the national crime Victimization survey used, created, or recoded

Survey item note

Teacher/adult notifieda

made fun of, called names

These items were used to create a new item for this study: “number of types of bullying experienced.”

Spread rumors

Threatened you

pushed, shoved, tripped, etc.

do things not wanted

excluded you

destroyed your property

how often happened this school year

School building

These items were used to create a new item for this study: “number of different locations bullying occurred.”

outside on school grounds

on a school bus

Somewhere else

none (physical injury)

These items were used to create a new item for this study: “Suffered physical injury.”

bruises or swelling

cuts, scratches

black eye/bloody nose

Teeth chipped/knocked out

broken bones/internal injuries

Knocked unconscious

other (physical injury)

a. Used as dependent variable in the analysis.

Source: U.S. Department of Justice 2007.


Table b2

bullying victim items from the 2007 national crime Victimization survey and the school crime supplement used, created, or recoded

Survey item note

gender Taken from the national crime victimization Survey

race and hispanic origin Taken from the national crime victimization Survey

current grade

region where student household located Taken from the national crime victimization Survey

household income Taken from the national crime victimization Survey

grades recoded into satisfactory grades (a’s, b’s, and c’s) and unsatisfactory grades (d’s and f’s) for this study

Skipped classes during academic year

adult at school who cares about me

School has adult that helps with problems

have friend at school to talk to

friend at school helps with problems

during school year in a fight

did you ever bring: gun

did you ever bring: knife as weapon These items were used to create a new item for this study: “brought weapon into school.”

did you ever bring: other weapon

how often student afraid someone will attack or harm them at school

how often student afraid someone will attack or harm them on way to/from school

besides school, how often student afraid someone will attack or harm them

Stay away from shortest route to school

Stay away from entrance to school

Stay away from hallway or stairs These items were used to create a new item for this study: “Student fear of attack and avoidance

Stay away from school cafeteria of school areas or activities scale.”

Stay away from restrooms

Stay away from other places inside school

Stay away from school parking lot

Stay away from other places on school grounds

avoid activities: attack harm you

avoid classes: attack harm you

Stay home: Thought someone attack harm you



Table b3

schools of bullying victim items from the 2007 school crime supplement to the national crime Victimization survey used, created, or recoded

Survey item note

School public or private

attend church-related school

everyone knows school rules

Same punishment for breaking the rules

School rules are fair

Students know punishments

School rules are strictly enforced

Teachers care about students

Teachers treat students with respect

Teachers make students feel bad

how often distracted by students misbehaving

how often teachers punish students for misbehaving

during school year know students on drugs/alcohol

during school year someone offered student illegal drugs/alcohol

possible to get alcohol

possible to get marijuana

possible to get prescription drugs

possible to get crack

possible to get cocaine

possible to get uppers

possible to get downers

possible to get lSd These items were used to create a new item for this study: “School crime and drug problem scale.”

possible to get pcp

possible to get heroin

Seen hate-related words or symbols

Know students brought gun to school

Seen student with gun

gotten a loaded gun

gangs at school

gangs involved in fights/violence at school

gangs sell drugs at school

School safety: security guards

School safety: staff/adults in hallway

School safety: metal detectors

School safety: locked doors

School safety: visitors sign in

School safety: locker checks

School safety: safety badges

School safety: security cameras

School safety: code of conduct



Table b4

unused items from the 2007 school crime supplement to the national crime Victimization survey

Survey item note

person number

These items are administrative and reflect

Total number of incidents for that person

first occurrence of this household

first occurrence of a person within this household

respondent line number

reason for ScS non-interview codes used by the interviewer.

incident start column location

incident record length

four digit year

adult present during questions

attend school this year These items reflect the screening criteria. Students home schooled who did not attend school at least part of the year were not included in the supplement. all or some home school

home school grade equivalent

Whether school was assigned or family chose the school no clear rationale for including in a study of reporting.

month current school begins Whether a student begins in august or September does not seem relevant to bullying.

lowest grade in school because grade level was already being analyzed, the specific grades included in the school were not analyzed. highest grade in school

respondent age This item is highly correlated with grade level (r = .906)

future: 4 year college This item is a follow-up item to the preceding item about attending school after high school.

future plans after high school no clear rationale for including in a study of reporting

participation in extracurricular activities no clear rationale for including in a study of reporting

how many days skipped classes This item is a follow-up item to the more inclusive question asking whether the student skipped any classes during the academic year.

number of times in a fight This item is a follow-up to the more inclusive item asking if the student was involved in any fights during the academic year.

harass: post This set of items is used by the national center for education Statistics to analyze cyber-bullying, harass: contact

harass: contact text although the items are described as “harassment.” it could not be determined whether these items are harassment how often already captured by the earlier bullying questions.

harassment: notify

hate related words: race

These questions are follow-up items asked if a student indicated “yes” to whether they saw hate-related words or symbols during the past academic year.

hate related words: religion

hate related words: ethnicity

hate related words: disability

hate related words: gender

hate related words: Sexual orientation

Whether student as ever called a hate related word at school no clear rationale for including in a study of reporting.

(conTinued)


Table b4 (conTinued)

unused items from the 2007 school crime supplement to the national crime Victimization survey

Survey item note

how long it took the student to get to school no clear rationale for including in a study of reporting.

how the student got to and from school no clear rationale for including in a study of reporting.

Whether students were allowed to leave school for lunch no clear rationale for including in a study of reporting.

how often students left school for lunch no clear rationale for including in a study of reporting.

future plans for school after high school no clear rationale for including in a study of reporting.

extra-curricular: athletics no clear rationale for including in a study of reporting.

extra-curricular: Spirit groups, pep no clear rationale for including in a study of reporting.

extra-curricular: arts no clear rationale for including in a study of reporting.

extra-curricular: academics no clear rationale for including in a study of reporting.

extra-curricular: Student government no clear rationale for including in a study of reporting.

extra-curricular: Service clubs no clear rationale for including in a study of reporting.

been called hate related words no clear rationale for including in a study of reporting.

assigned school or family choose no clear rationale for including in a study of reporting.

how long to school no clear rationale for including in a study of reporting.

how get to school no clear rationale for including in a study of reporting.

how get home from school no clear rationale for including in a study of reporting.

Students allowed to leave school at lunch no clear rationale for including in a study of reporting



Center for Education Statistics website (www.nces. ed.gov/quicktables) following each survey data release. The survey data have been used extensively in research studies and reports. For example, DeVoe and Kaffenberger (2005) used survey data to examine victim and school characteristics of students who were victims of direct and indirect bullying behaviors. However, to date, no National Center for Education Statistics publications have used the survey to specifically examine the reporting of bullying victimization to school officials.

Identifying reported and unreported bullying

To first identify whether students responding to the National Crime Victimization Survey School Crime Supplement were bullied, interviewers stated the following: “Now I have some questions about what students do at school that make you feel bad or are hurtful to you. We often refer to this as being bullied” (U.S. Department of Justice 2009). Students were considered bullied if they responded affirmatively to questions that probed whether they were bullied in one or more of the following ways: being made fun of; being the subject of rumors; being threatened with harm; being pushed, shoved, tripped, or spit on; being made to do things they did not want to do; being excluded from activities on purpose; and having property destroyed on purpose. About 32 percent of students in the 2007 survey indicated they had been the victims of at least one type of bullying behavior during the last academic year.

To identify whether bullied students indicated whether their victimization was reported to a teacher or other adult at the school, students were asked, “Was a teacher or some other adult at the school notified about (this event/any of these events)?” Of the roughly 32 percent of students who reported at least one bullying incident on the survey, 36 percent reported that their victimization was reported to a school official and 64 percent did not.5 The question does not permit the researchers to identify who reported the bullying victimization, be it the student, a parent, or someone else (such as a bystander).

Handling survey nonresponse and complex survey sampling

Two major issues about the way the sample was constructed were taken into account. First, not all students eligible to respond to the survey participated, which could bias results if those who responded are different in substantive ways from those who did not. According to the Bureau of Justice Statistics (U.S. Department of Justice 2007), nonparticipating students are more likely to come from non-White, urban, and lower income households. Therefore, a person weight is used to take nonresponse into account and to provide more accurate estimates of population parameters.6

Weighting helps account for potential biases due to nonresponse and permits inferences from these data to the national population of student bullying victims. The Bureau of Justice Statistics (2007, p. 172) concludes that, “although the extent of non-response bias cannot be determined, weighting adjustments, which corrected for differential response rates, should have reduced the problem.”

All tests of statistical significance were based on unweighted sample sizes, but the descriptive results (the percentages of reported and nonreported bullying victimization) were weighted to provide national population estimates. So although 1,778 students indicated that they were bullied during the previous academic year (the total number of observations), the weighted estimates reported are based on 7,775,000 students and represent a national estimate of student bullying victims, a procedure the National Center for Education Statistics follows when reporting these and other nationally representative survey data (Bauer et al. 2008; Dinkes, Kemp, and Baum 2009).

Second, the survey uses a stratified, multistage cluster sample design. Analyzing such data without taking this complexity of sampling into account could result in biased estimates. The Bureau of Justice Statistics (2007) recommends that standard errors be computed in a manner that takes this type of sampling into account. The complex sampling design used for the survey


required sample weights, sampling units (clusters), and sampling strata to adjust for clustering and stratification to compute valid standard errors. The Stata 11 statistical package (StataCorp 2009) was used, and the analysis was conducted using the Taylor series linearization method with primary sampling units and strata variables available in the data set.

Handling item nonresponse and proxy interviews

Two other methodological issues, apart from the sampling, also required attention: item non-response and proxy interviews. Item response rates were 95–99 percent for nearly all items in the analysis, meaning that there is little potential for item nonresponse bias in the results. Table B5 lists items with response rates below 95 percent. The potential for bias still exists for these variables, so analysis involving them should be interpreted with caution. This is the standard used by the National Center for Education Statistics when analyzing these same data.7

Allison (2002) argues that when the percentage of item data missing is low (a few percent of missing cases), complete case analysis can be done—that is, analysis can be conducted only on cases for which all data are available—with no

concern for error. This is also known as “listwise deletion.” And even if item nonresponse is 15 percent or higher, weighted adjustments that address survey nonresponse may also reduce the problem of item nonresponse bias (Bauer et al. 2008).

Second, for a few interviews a parent or other guardian in the household provided the data by proxy for the student. Sensitivity analysis was conducted to determine whether including or excluding the proxy interviews changed the findings (table B6). Proxy interviews comprised such a small percentage (2.3 percent) of student bullying victims that their impact on the overall analyses was negligible. The results without the proxy interviews indicate marginal changes in the overall percentages and no changes in the results of significant tests for the variables.

“ ’ ”

Table b5

survey items with less than 95 percent response rate

item

response rate

(percent)

response rate with

don t know as missing

School safety: locked doors 100.0 94.9

how often this happened this school year (bullying) 99.8 94.9

future after high school 99.5 94.2

School safety: locker checks 100.0 94.0

School safety: security cameras 100.0 89.8

household income 80.2 80.2


Selecting items

As mentioned, the National Crime Victimization Survey School Crime Supplement has approximately 140 items. This project was designed to be descriptive, and it is not uncommon for such projects to analyze a large number of variables. Similar research studies using large, national survey data sets, including National Center for Education Statistics reports on school crime and safety, have reported on large numbers of variables. For example, the Nieman and DeVoe (2009) study using data from the School Survey on Crime and Safety includes separate analyses of nearly 100. This study included 51 items from the survey about the student’s bullying victimization, the individual student, and the school that the student attends. A few selected items from the household portion of the larger National Crime Victimization Survey, such as household income and region, were also included. See tables B1–B4 for further information on the items that were used, created, or recoded for the analysis. Whether a student was bullied was used to define the subpopulation of bullying victims, and the “teacher/adult notified” variable was the dependent variable in the cross-tabulations.

Table b6

Relationship between bullying characteristics and reporting of bullying,during the 2007 school year, without proxy interviews

as indicated by students ages 12–18




error -chi square

value

chance probability

of result

victim of bullying

victim of bullying 1,736 35.7 64.4 — — —

physical injury to victim

yes 115 59.1 40.9 5.34 31.6* <.001

Types of bullying

Threatened 313 54.5 45.5 2.96 65.95* <.001

destroyed property 230 51.5 48.5 3.51 32.47* <.001

pushed, shoved, tripped, and the like 611 45.0 55.0 2.11 37.70* <.001

made fun of, called names 1,145 37.9 62.1 1.42 8.38 .005

excluded 295 37.8 62.2 2.99 .763 .458

Spread rumors 991 37.2 62.8 1.59 2.64 .152

victim made to do things he or she did not want to do 232 36.9 63.1 3.21 .200 .674

number of types of bullying experienced

one 697 25.7 74.3 1.75

70.9* <.001

Two 473 37.1 62.9 2.43

Three 283 44.7 55.3 3.22

four 157 44.8 55.2 3.71

five 78 52.7 47.3 5.49

Six 31 59.4 40.6 8.61

Seven 17 50.3 49.7 12.16

frequency of bullying during academic year

once or twice this school year 1,029 32.1 67.9 1.65

23.97* <.001 once or twice a month 344 39.7 60.3 2.37

once or twice a week 168 44.3 55.7 3.47

almost every day 111 49.1 50.9 4.84

location where bullying occurred

School building 1,371 36.4 63.6 2.30 1.98 .144

outside on school grounds 394 38.1 61.9 2.35 1.49 .232

School bus 145 47.5 52.5 4.25 10.29* .002

Somewhere else 68 25.8 74.2 4.91 3.23 .058

number of different locations bullying occurred

one 1,522 34.6 65.4 1.27 11.04* .002

Two or more 197 46.1 53.9 3.72

— is not applicable.

* Difference between bullying characteristic and reporting is statistically significant, p < .0033.




There are 15 items within the bullying victimization domain. Because of the small number of bullying victims that experienced each type of physical injury, a single injury item was created by collapsing all physical injury responses (for example, “cuts, scratches,” “bruises, swelling”) together. All seven types of bullying (for example, “being excluded” or “being called names”) were analyzed and used to create an item indicating how many different types of bullying a victim experienced (ranging from one to seven). The frequency and location variables were taken from the Student Crime Supplement, and the location variables were also used to create a new item indicating whether a student was victimized in one or multiple locations. According to Unnever and Cornell (2004), items focused on the severity and frequency of bullying are most relevant to whether bullying is reported.

There are 14 items within the student victim domain. Sociodemographic characteristics (gender, race/ethnicity, and household income) are included because they are routinely analyzed in studies with national survey data (for example, Dinkes, Cataldi, and Lin-Kelly 2008). Because the National Crime Victimization Survey School Crime Supplement is a national survey, an item indicating the region of the country in which the student’s household is located is included to determine whether there are differences between the Census’ Northeast region (which overlaps substantially with Regional Educational Laboratory Northeast and Islands) and other regions of the United States. Prior research (for example, Middleton 2008) indicates the negative relationship of grade level to reporting, so this variable is also included. DeVoe and Kaffenberger (2005) have studied the relationship between academic performance and bullying itself but not between academic performance and reporting of bullying, so academic performance and whether the student has skipped school are included. Four items relevant to “protective factors”—namely relationships with adults and friends at the school—are also included. Students that have an adult at school who cares about them (Benard

2004) or who helps them solve problems may be more likely to disclose being bullied. Students that have a friend whom they can talk to or who helps them solve problems may be less likely to report to a school official. Because the frequency of weapon carrying is very low, the survey items whether a student brought a gun, knife as weapon, or other weapon to school were collapsed into a single item “brought weapon to school.” Whether students were involved in a fight or brought a weapon to school may reflect students’ willingness to protect themselves physically or to personally “settle the score” and not report their victimization to a school official. Finally, the student victim domain also includes a scale based on how fearful the student was of being attacked and on whether the student avoids certain school areas or activities. As fear and avoidance increase, bullying victims may be more reluctant to come forward to report their victimization (Oliver and Candappa 2007).

There are 22 items in the school domain. Items that indicated whether the school was public or private, or church-affiliated, were analyzed to determine whether the reporting of bullying varied by the school’s structural characteristics. Ten items examine school culture and classroom environment. Unnever and Cornell (2004) found that students were more likely to report their bullying if they perceived that the school’s culture was not tolerant of bullying. Other research indicates that schools in which students feel positive toward their school and teachers and schools in which classrooms have few disruptions due to behavioral issues are less likely to have a bullying problem (Swearer et al. 2010). Crime, drug, and bias incidents may signal to students that their school is dangerous and disorderly, which could affect reporting. The school crime and drug problem scale was created by combining 13 items related to those factors. Schools are implementing a variety of security measures (such as metal detectors), so nine items related to security measures in the school were also included.

Some items were not used because they did not have clear, justifiable rationale for inclusion in

29

a project about the reporting of bullying (see table B4). Others that were not analyzed included administrative variables used by Census Bureau interviewers (such as respondent line number) or screening variables used to remove ineligible household members from the School Crime Supplement (such as whether the respondent attended school this year). In a few instances, only the first in a series of items was analyzed. For example, whether the student skipped school during the academic year was analyzed, but the number of days school was skipped was not. In another instance, age was found to be highly correlated with grade level (r = .906) for a sample that only involves students, ages 12–18 and in grades 6–12, so the age variable was not included.

For some items, response categories were collapsed for the analysis. For the most part, this involved items that asked students questions about their level of agreement with a statement. For example, students were provided a statement “School rules are fair,” and asked whether they “agree,” “strongly agree,” “disagree,” or “strongly disagree” with the statement. “Agree” and “strongly agree” responses were combined, as were “disagree” and “strongly disagree” responses.

Scaled items

As mentioned in the previous section, two scales—a student fear of attack and avoidance of school areas or activities scale and a school crime and drug problem scale—were created to simplify the analysis because several items in the National Crime Victimization Survey School Crime Supplement ask about the same underlying construct, and there appeared to be limited value to analyzing and presenting results for the individual items comprising the scales.

For the student fear of attack and avoidance of school areas or activities scale, students were asked three items about their fear at school, their fear on the way to or from school, and their fear about being attacked or harmed outside of school. For these three items, students indicated whether

appendix b. daTa Source and meThodology

they were never afraid, almost never afraid, sometimes afraid, and afraid most of the time. “Never afraid” and “almost never afraid” responses counted for 0 points on the scale, and “sometimes afraid” and “afraid most of the time” responses counted for 1 point on the scale. Students were then asked 11 items about whether they avoided school, certain activities at school, or certain locations in the school because of their fear of attack. Each location or activity that a student avoided because of fear of attack counted for 1 point on the scale.

The school crime and drug problem scale measured whether the student knew other students who brought a loaded gun to school, whether he or she had seen another student with a loaded gun at school, whether he or she could have acquired a loaded gun at school, whether gangs were at school, whether gangs were involved in selling drugs at school, whether gangs at school were involved in fights and violence, whether he or she had seen hate-related words and symbols at school, whether he or she was offered drugs or alcohol during the academic year, whether he or she knew other students on drugs or alcohol, whether it was possible to get alcohol at school, whether it was possible to get marijuana at school, whether it was possible to get prescription drugs at school, and whether it was possible to get crack, cocaine, or other drugs at school. Each “yes” response counted for 1 point on the scale.

Because both scales comprised yes or no (0 or 1) responses to individual items, the Kuder-Richardson coefficient of reliability of the individual items was computed. Some researchers advocate minimum reliability coefficients with dichotomous data between .70 and .80 (Netemeyer, Bearden, and Sharma 2003; Robinson, Shaver, and Wrightsman 1991). When rounded, both reliability coefficients are .80 or higher. The reliability coefficients were computed on the sample of bullying victims rather than the entire sample. A pointbiserial correlation was calculated between each scale and reporting of bullying. The student fear of attack and avoidance of school areas and activities

30 characteristics of bullying, bullying victims, and schools associated with reporting of bullying

scale met the Bonferroni adjusted significance level of .0036, confirming the chi-square result in table 3 (figure B1); the reliability coefficient for the scale is .80 (table B7). A point-biserial correlation was calculated between the school crime and drug problem scale and reporting. The school crime anddrug problem scale did not meet the Bonferroni adjusted significance level of .0023, confirming the chi-square result in table 5 (figure B2); the reliabil-ity coefficient for the scale is .84 (table B8).

Chi-square analysis

Descriptive analysis (cross-tabulations) was conducted to respond to the research questions, focusing on comparisons between reported and unreported bullying according to self-reports by bullying victims. Cross-tabulations were usually composed of 2×2 tables analyzing the presence or absence of a characteristic with reporting or

figure b1 Point-biserial correlation between the student fear of attack and avoidance of school areas and activities scale and reporting

. pbis vs087 fearavoid2 if vr1==1(obs = 1772)Np = 625 p = 0.35Nq = 1147 q = 0.65

Coef. = 0.1150 t = 4.8722 P>|t| = 0.0001 df = 1770

Source: Authors’ calculations based on U.S. Department of Justice 2007.

figure b2 Point-biserial correlation between the school crime and drug problem scale and reporting

. pbis vs087 schcrime(obs = 1158)Np = 429 p = 0.37Nq = 729 q = 0.63

Coef. = –0.0623 t = –2.1224 P>|t| = 0.0340 df = 1156


table b7 Kuder-Richardson coefficient of reliability factivities scale

or the student fear of attack and avoidance of school areas and

number of items in the scale = 14number of complete observations = 1,782

number of item observations item difficulty item variance item-test correlation

vs106 1,782 0.9444 0.0525 0.4088

vs107 1,782 0.9736 0.0257 0.4303

vs108 1,782 0.9456 0.0515 0.5066

vs109 1,782 0.9551 0.0429 0.4520

vs110 1,782 0.9506 0.0469 0.4587

vs111 1,782 0.9691 0.0299 0.4350

vs112 1,782 0.9703 0.0289 0.3175

vs113 1,782 0.9736 0.0257 0.3756

vs114 1,782 0.9590 0.0393 0.4054

vs115 1,782 0.9826 0.0171 0.4179

vs116 1,782 0.9815 0.0182 0.3263

fearattrecode1 1,782 0.9052 0.0858 0.5008

fearattrecode2 1,782 0.9540 0.0439 0.4260

fearattrecode3 1,782 0.9383 0.0579 0.4240

test 0.9574 0.4203

Kr20 coefficient is 0.7976.


Table b8

Kuder-Richardson coefficient of reliability for the school crime and drug problem scale

number of items in the scale = 13 number of complete observations = 1,165

number of item observations item difficulty item variance -item test correlation

vs123 1,165 0.0961 0.0869 0.2832

vs124 1,165 0.0343 0.0332 0.2428

vs125 1,165 0.0987 0.0890 0.3651

vs126 1,165 0.2601 0.1924 0.5078

vs128 1,165 0.1330 0.1153 0.6099

vr16 1,165 0.3056 0.2122 0.6697

alcoholrecode 1,165 0.2524 0.1887 0.5585

marijuanarecode 1,165 0.3674 0.2324 0.7052

prescriptionrecode 1,165 0.2893 0.2056 0.6116

ganginfightsviolenceSchrecode 1,165 0.2000 0.1600 0.5288

vs105 1,165 0.5142 0.2498 0.2832

vs066 1,165 0.4918 0.2499 0.5413

vs067 1,165 0.1502 0.1277 0.5032

Test 0.2456 0.4931

Kr20 coefficient is 0.8433.



nonreporting), but some larger tables were also used.

To test for differences between reporters and nonreporters, Pearson’s chi-square was used because the variables were categorical in nature. Chi-square analysis is a statistical technique that measures the discrepancy between the observed cell counts and what would be expected if the rows and columns were unrelated. If the rows and columns are related (that is, if the chi-square test shows a statistically significant result by the standards explained below), the characteristic is found to be related to or associated with the independent variable (in this case, reporting). In short, chi-square analysis indicates whether there are significant variations in the distribution of a particular characteristic between reported and nonreported bullying victimization. Because 35.8 percent of the total bullying victim sample indicated their victimization was reported, a chi-square will be more likely to be statistically significant the more

the prevalence of reporting along a particular variable (such as gender) departs from this overall sample finding.

Since chi-square analysis does not indicate the direction of the relationship of two variables, correlations were calculated for statistically significant items to determine whether a variable was associated with an increase or decrease in reporting. Point-biserial correlations (rpbi) were used to indicate directionality for the two scales (student fear of attack and avoidance of school areas or activities scale and school crime and drug problem scale) and other continuous variables. For statistically significant dichotomous variables, tetrachoric correlations (rho, appropriate for 2×2 tables of categorical data) are reported to indicate directionality (Welkowitz, Ewen, and Cohen 1982). All correlations procedures have similar qualities in that they range from 0 to 1, with 0 indicating no relationship and 1 indicating perfect relationship. In addition, the correlations can be positive


or negative to indicate the directionality of the relationship (Welkowitz, Ewen, and Cohen 1982).

The initial threshold to determine statistical significance was set at p = .05 (two-tailed). But because of the number of significance tests conducted, there is an increased likelihood of some results being statistically significant due to chance. To guard against this, a Bonferroni multiple comparison procedure was calculated to adjust for the number of significance tests (Bland and Altman 1995). Specifically, the critical value of the significance test (0.05) was divided by the number of statistical tests calculated within each of the three research domains: 15 analyses were conducted in the characteristics of bullying victimization domain (research question 1), 14 analyses in the characteristics of bullying victims domain (research question 2), and 22 analyses in the characteristics of bullying victims’ schools domain (research question 3). The Bonferroni procedure yields adjusted statistical significance levels of 0.0033 for characteristics of bullying victimization, 0.0036 for characteristics of bullying victims, and 0.0023 for characteristics of bullying victims’ schools. These adjusted levels are used to identify statistically significant associations.

Regardless of whether the findings are statistically significant, it is important to note that all the data analysis is descriptive and does not allow for causal interpretation. No conclusions about the effectiveness of school policies and strategies on the reporting of bullying can be reached.

Further limitations of the study

The National Crime Victimization Survey School Crime Supplement was designed to present data on a wide range of crime, safety, and discipline issues in schools; it is not focused specifically on bullying. It contains only one item on whether the bullying experienced was reported to an adult, and that item is not linked to any specific bullying incident or time sequence, so it cannot be determined whether reporting occurred after a specific type of bullying or after a specific amount of time.

The survey also relies on respondents to self-determine their condition as a victim of bullying using their own interpretation and conceptions to define a situation as bullying. Although this is considered an improvement over official reports because bullying victims are often reluctant to report victimization to school officials, self-reports are susceptible to other biases (Unnever and Cornell 2004). Similar situations may not be labeled as bullying by different respondents. The survey does not specifically mention how often victimization has to be repeated to be defined as bullying. Students are asked how persistent the bullying has been, and responses can range from “once or twice this school year” to “nearly every day.” The National Center for Education Statistics includes all students who have experienced bullying as bullying victims regardless of how often the bullying occurred, a position also taken for this study. Some students may be reluctant to tell an interviewer about being bullied, so some victims may not be reported as such. Although research conducted across 14 countries by Smith et al. (2002) indicates that children are able to differentiate bullying from teasing and other behaviors, the extent of bullying misspecification in the National Crime Victimization Survey School Crime Supplement is unknown. Ideally, other measures related to bullying status and whether it was reported would be derived from independent observation or other means. But having variables that represent students’ interpretation of their surroundings that may influence reporting, which is of primary interest to this investigation, may be viewed as a strength. No attempt was made by the survey researchers or by the research team to determine whether students correctly self-reported their bullying victimization, their reporting to school officials, or any other information they provided interviewers.

The states in which students reside are not identified in the publicly available survey data, so it is not possible to provide more fine-grained analysis by jurisdiction. However, analysis conducted using the national sample take advantage of the statistical power provided by the increased sample size. The data do permit classification by Census region


(“Northeast” comprises the seven Regional Educational Laboratory Northeast and Islands states— Connecticut, Maine, Massachusetts, New Hampshire, New York, Rhode Island, and Vermont—as well as Pennsylvania and New Jersey, but does not include the U.S. Virgin Islands or Puerto Rico. Table 2 shows the results of one analysis that compares the Northeast region with other regions.

The survey results obtained from these data would have more credibility if validated by other research. The data on whether bullying victims reported their victimization to school officials cannot be verified independently to determine the accuracy of the estimates, but another way to validate survey findings is to determine whether the estimates reported in this project are “reasonable.” To do this, other U.S. national survey results that include items on bullying and reporting to school officials are needed. Although there are a few national surveys that include an item about bullying (such as the Youth Risk Behavior Surveillance System), none that includes an item about the reporting of bullying victimization by student victims to school officials could be found. Middleton (2008) included a question on reporting, but that study covered Oklahoma only and did not indicate the time frame for which students were asked to recall their victimization (the School Crime Supplement asks students to indicate whether they were victimized during the past academic year). The age groups covered by the survey are also different (the School Crime Supplement does not cover students in grades 3 and 5, so only students in grade 7 could be compared). The Unnever and Cornell (2004) study, which was a more intensive investigation of reporting, includes only middle schools (grades 6–8) in Roanoke, Virginia.

The cross-tabulations conducted for this study consist of descriptive analysis of the relationship of one variable (within the three domains of bullying victimization, bullying victim, or bullying victims’ schools) with another (reporting or nonreporting). Such simple cross-tabulations do not control for additional variables, as could be done, for example, in a more advanced multivariate statistical analysis.

The data analysis is also limited to the variables available in the data set. Other unmeasured variables that may influence reporting behavior cannot be accounted or controlled for. Moreover, the instrument does not ask students whether they witnessed bullying and reported it. Bystander nonreporting in bullying is considered a critical ingredient to the “culture of silence” in schools (see, for example, Hendricks 2008). There are also no data on bullying perpetrators, and victims are not asked about the characteristics of their victimizers in the National Crime Victimization Survey School Crime Supplement. Unnever and Cornell (2004) were able to analyze the “perceived toleration of bullying at the school” in their analysis of reporting versus unreported bullying in Roanoke, Virginia, but the National Crime Victimization Survey School Crime Supplement does not include such items in its questionnaire.

Finally, sampling error presents another limitation. Because the sample of students selected for each administration of the School Crime Supplement is just one of many possible samples that could have been selected, it is possible that estimates from a given sample may differ from estimates that would have been produced from other randomly drawn student samples.


noTes Psychologists, the National Association of Secondary School Principals, Fight Crime: Invest

The authors thank the following people for their in Kids, the Learning Disabilities Association comments and assistance at various stages of the of America, and the National PTA. project: Candice Bocala, Rebecca Carey, Katie Culp, David Farrington, Craig Hoyle, Kevin 5. Approximately 0.6 percent of students who Huang, Jo Louie, Susan Mundry, Laura O’Dwyer, were bullied did not respond to the item about Dan Oleweus, OK-Choon Park, David Perda, Jan whether the bullying incident was reported to Phlegar, Mary Stenson, Maria Ttofi, Jill Weber, a teacher or some other adult at the school. and the ATS peer reviewers.

6. Weights are numbers added or accumulated 1. It could not be determined whether the U.S. to obtain universe estimates of particular

Virgin Islands is considering anti-bullying events. The final weight is a multiplier that legislation at the time of writing. indicates how many times a particular sample

record is to be counted (U.S. Department of 2. Personal correspondence, Lourdes Rivera- Justice, Office of Justice Programs, Bureau of

Putz, December 2, 2008. Justice Statistics 2007).

3. Personal correspondence, Lourdes Rivera 7. National Center for Education Statistics (2010) Putz, December 2, 2008. Statistical Standard 2-2-2-4 states that, “If the

item response rate is below 85 percent for any 4. Organizations that are part of the coalition items used in a report, a nonresponse bias

include the American Federation of Teach- analysis is also required for each of those items ers, the National Association of School (this does not include individual test items).”

35 referenceS

RefeRences

Allison, P.D. (2002). Missing data. Thousand Oaks, CA: Sage.

Associated Press. (2009, September 14). School bullying laws give scant protection. USA Today. Retrieved from http://www.usatoday.com/news/education/2009-09-14 -bullying-laws_N.htm

Bauer, L., Guerino, P., Nolie, K.L., Tang, S.W., and Chandler, K. (2008). Student victimization in U.S. schools: results from the 2005 school crime supplement to the National Crime Victimization Survey. Washington, DC: U.S. Department of Education, Institute of Education Sciences.

Benard, B. (2004). Resiliency: what have we learned. San Francisco, CA: WestEd.

Bland, J.M. and Altman, D. (1995). Multiple significance tests: the Bonferroni method. Statistics Notes. British Medical Journal, 310, 170.

Boorstein, M. (2004, November 7). In suit, VA teen accuses schoolmates of bullying. The Washington Post, p. C1.

Canadian Psychological Association. (2009). Policy statements: bullying in children and youth. Retrieved from http://www.cpa.ca/aboutcpa/policystatements/#Bullying.

CNN (2001, March 8). Study: kids rate bullying and teasing as big problem. Retrieved from http://archives.cnn. com/2001/US/03/08/violence.survey/.

Dawson, D. (2006, July 31). Girls take school to court, saying it ignored bullying. Retrieved from http://abcnews. go.com/US/LegalCenter/story?id=2256089andpage=1.

DeVoe, J.F., and Kaffenberger, S.K. (2005). Student reports of bullying: results from the 2001 School Crime Supplement to the National Crime Victimization Survey (NCES 2005-310). Washington, DC: U.S. Department of Education.

Dinkes, R., Cataldi, E.F., and Lin-Kelly, W. (2008). Indicators of school crime and school safety: 2007. Washington, DC: U.S. Department of Education, Institute of

Education Sciences, National Center for Education Statistics.

Dinkes, R., Kemp, J., and Baum, K. (2009). Indicators of school crime and safety: 2008 (NCES 2009–022/NCJ 226343). Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, and U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Education Development Center (2008). Changing the rules of the game. EDC Update Spring, 3–4.

Farrington, D.P., and Ttofi, M.M. (2009). Reducing school bullying: evidence-based implications for policy. Crime and Justice: An Annual Review of Research, 38, 281–345.

Fekkes, M., Pijpers, F., and Verloove-Vanhorick, S. (2004). Bullying behavior and associations with psychosomatic complaints and depression in victims. Journal of Pediatrics, 144(1), 17–22.

Fekkes, M., Pijpers, F., and Verloove-Vanhorick, S. (2005). Bullying: who does what, when and where? Involvement of children, teachers and parents in bullying behavior. Health Education Research, 20(1), 81–91.

Fosse, G.K., and Holen, A. (2002). Childhood environment of adult psychiatric outpatients in Norway having been bullied in school. Child Abuse and Neglect, 26(2), 129–137.

Fox, J. A., Elliot, D., Kerlikowske, R.G., Newman, S., and Christeson, W. (2003). Bullying prevention is crime prevention. Washington, DC: Fight Crime: Invest in Kids.

Garbarino, J., and DeLara, E. (2003). And words can hurt forever: how to protect adolescents from bullying, harassment and emotional violence. New York: Simon and Schuster.

Glew, G.M., Fan, M.Y., Keaton, W., Rivera, F.P., and Kernic, M.A. (2005). Bullying, psychosocial adjustment, and academic performance in elementary school. Archives of Pediatric and Adolescent Medicine, 159(11), 1026–1031.


Griffin, R.S., and Gross, A.M. (2004). Childhood bullying: current empirical findings and future directions for research. Aggression and Violent Behavior 9, 379–400.

Hagan, F.E. (1993). Research methods in criminal justice and criminology. Third edition. New York: MacMillan.

Halligan, J. (2005, August 17). Death by cyber-bully. Boston Globe. Retrieved from http://www.boston.com/news/ globe/editorial_opinion/oped/articles/2005/08/17/ death_by_cyber_bully/.

Hendricks, L. (2008, December 22). Schools target bullies. The Daily News (Newburyport, MA). Retrieved from http://www.newburyportnews.com/punews/local_ story_356234448.html.

Hergert, L. (2004). Bullying and students with disabilities: summary report of parent focus groups. Newton, MA: Education Development Center. Retrieved from http:// www.urbancollaborative.org/pdfs/Bullying.pdf.

Juvonen, J., Graham, S., and Schuster, M.A. (2003). Bullying among young adolescents: the strong, the weak, and the troubled. Pediatrics 112(6), 1231–1237.

Kaiser Family Foundation. (2001). Talking with kids about tough issues: a national survey of parents and kids. Retrieved June 3, 2010, from http://www.talkwithkids. org/nickelodeon/details.pdf.

Kazdin, A. and Rotella, C. (2009, August 11). Bullies: they can be stopped, but it takes a village. Slate. Retrieved March 3, 2010, from http://www.slate.com/id/2249424.

King, L., and Hendricks, L. (2010, February 8). Local schools work to combat bullying: recent incidents serve to highlight ongoing issue. The Daily News (Newburyport, MA). Retrieved from http://www.newburyportnews. com/punews/local_story_038222912.html.

Klass, P. (2009, June 9). At last, facing down bullies (and their enablers). New York Times, p. D-5.

Larkin, D. (2007). Vermont girls report facing bullying, harassment. Retrieved from http://www.timesargus.com/apps/ pbcs.dll/article?AID=/20070321/NEWS01/703210355/1002.

Maine Governor’s Children’s Cabinet. (2006). Maine’s best practices bullying and harassment prevention: a guide for schools and communities. Augusta, ME: Maine Governor’s Children’s Cabinet.

Marshall, M.P. (2010, March 9). Students get a message: no bullying at Pine Grove. The Daily News (Newburyport, MA). Retrieved from http://www.newburyportnews. com/punews/local_story_067232804.html/ resources_etp_mobile_story.

Martindale, S. (2009, June 1). Parents of teen who shot himself file $3 million claim against district. OC [Orange County] Register. Retrieved from http://articles.latimes.com/2009/jun/04/local/ me-suicide-bully4.

Middleton, K. (2008). Bullying perceptions of third, fifth and seventh grade students in Oklahoma public schools, 2005. Oklahoma City: Oklahoma Department of Health.

National Crime Prevention Council. (2008). Bullying. Retrieved from http://www.ncpc.org/topics/bullying.

National Institutes of Health. (2001, April 24). Bullying widespread in U.S. schools, survey finds. NIH News Release. Retrieved from http://www.nichd.nih.gov/news/ releases/bullying.cfm.

National Safe Schools Partnership. (2007). Bridging the gap in federal law: promoting safe schools and improved student achievement by preventing bullying and harassment in our schools. Washington, DC: National Safe Schools Partnership. Retrieved from https://sites.nea. org/schoolsafety/images/bridginggap.pdf.

Neiman, S., and DeVoe, J.F. (2009). Crime, violence, discipline, and safety in U.S. public schools: findings from the school survey on crime and safety: 2007–08 (NCES 2009-326). Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics.

Netemeyer, R.G., Bearden, W.O., and Sharma, S. (2003). Scaling procedures: issues and applications. Thousand Oaks, CA: Sage Publications.

37 referenceS

Oliver, C., and Candappa, M. (2007). Bullying and the politics of telling. Oxford Review of Education, 33(1), 71–86.

Olweus, D. (1993). Bullying at school: what we know and what we can do. New York: Blackwell.

Parker-Roerdon, L., Rudewick, D., and Gorton, D. (2007). Direct from the field: a guide to bullying prevention. Boston, MA: Commonwealth of Massachusetts, Department of Public Health.

Pepler, D., and Craig, W. (Eds.). (2008). Understanding and addressing bullying: an international perspective. Bloomington, IN: Authorhouse.

Peterson, L. (2009, June 10). Officials hope anonymous online reports will reduce bullying. The Tampa Tribune. Retrieved from http://www2.tbo.com/content/2009/jun/10/ school-officials-hope-anonymous-online-reports-wil/.

Rigby, K. (2003). Consequences of bullying in schools. Canadian Journal of Psychiatry, 48(9), 583–590.

Robinson, J. P., Shaver, P. R., and Wrightsman, L. S. (1991). Criteria for scale selection and evaluation. In J. P. Robinson, P. R. Shaver, and L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 1–15). San Diego, CA: Academic Press.

Sampson, R. (2004). Bullying in schools: problem-oriented guides for police. Problem-Specific Guides Series. Guide No. 12. Washington, DC: Office of Community-Oriented Policing Services, U.S. Department of Justice.

Sidorowicz, K., Hair, E.C., and Milot, A. (2009). Assessing bullying: a guide for out-of-school program practitioners. Research-to-Results Brief, Publication No. 2009–42. Washington, DC: Child Trends.

Smith, P.K., Cowie, H., Olafsson, R.F., and Liefooghe, A. (2002). Definitions of bullying: a comparison of terms used, and age and gender differences, in a fourteen country international comparison. Child Development, 73, 1119–1133.

Sourander, A., Helstelä, L, Helenius, H., and Piha, J. (2000). Persistence of bullying from childhood to

adolescence—a longitudinal 8-year follow-up study. Child Abuse and Neglect 24(7), 873–881.

Sourander A., Jensen P, Ronning J.A., Elonheimo, H., Niemela, S., Helenius, H., et al. (2007) Childhood bullies and victims and their risk of criminality in late adolescence: the Finnish From a Boy to a Man study. Archives of Pediatric and Adolescent Medicine. 161, 546–552.

Srabstein, J., and Piazza, T. (2008). Public health, safety, and educational risks associated with bullying behaviors in American adolescents. International Journal of Adolescent Medicine and Health, 20(2), 223–233.

StataCorp. (2009). Stata users guide: release 11. College Station, TX: StataCorp LP.

Swearer, S.M., Espelage, D.L., Valliancourt, T., and Hymel, S. (2010). What can be done about school bullying? Linking research to educational practice. Educational Researcher 39(1), 38–48.

Teicher, S.A. (2006, October 19). How students can break the ‘code of silence.’ Christian Science Monitor, p. 15.

Unnever, J.D., and Cornell, D.G. (2004). Middle school victims of bullying: who reports being bullied? Aggressive Behavior 30, 373–388.

U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics. (2010). Statistical standards: planning and design of surveys. (NCES Standard 2-2). Retrieved from http://nces. ed.gov/statprog/2002/std2_2.asp.

U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. (2007). National Crime Victimization Survey, School Crime Supplement, 2005 [Codebook]. Ann Arbor, MI: Interuniversity Consortium for Political and Social Research.

U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. (2009). National Crime Victimization Survey, School Crime Supplement, 2007 [Codebook]. Ann Arbor, MI: Interuniversity Consortium for Political and Social Research.


Vaznis, J. (2009, November 15). Support swells for anti-bullying legislation. Boston Globe. Retrieved from http://www.boston.com/ news/education/k_12/articles/2009/11/15/ support_swells_for_anti_bully_legislation/?page=1.

Vossekull, B., Fein, R., Reddy, M., Borum, R., and Modzeleski, W. (2002). The final report and findings of the safe school initiative: implications for prevention of school attacks in the United States. Washington, DC: U.S. Secret Service and U.S. Department of Education.

Welkowitz, J., Ewen, R.B., and Cohen, J. (1982). Introductory statistics for the behavioral sciences. Third edition. San Diego, CA: Harcourt Brace Jovanovich.

The White House. (2009). Prepared remarks of President Barack Obama: Back to School Event, September 8, 2009, Arlington, Virginia. Retrieved December 17, 2009, from http://www.whitehouse.gov/ MediaResources/PreparedSchoolRemarks/

1

THE OCR LETTER



Ph. 847-564-8662 Fax 847-564-8419



Notes #4



2

OCR Letter Dated October 26, 2010 on Discriminatory Harassment

Introduction: The purpose of this letter, which was sent to all education executives, was to remind them that some acts of bullying and harassment may also fall within OCR jurisdiction.

I. Mandated Actions Highlights

Schools, when limiting their inquiry to policy and code of conduct violations, may improperly be limiting the federally required scope of their inquiry. The real standard is:

When peer harassment based on race, color, national origin, sex, or disability is sufficiently serious that it creates a hostile environment and such harassment is encouraged, tolerated, not adequately addressed, or ignored by school employees.

This definition also encompasses post secondary institutions. Harassment creates a hostile environment when the conduct is sufficiently severe, pervasive or persistent so as to limit or interfere with a student’s ability to participate in or benefit from the services, activities, or opportunities offered by a school. A school is responsible for addressing harassment incidents about which it knows or reasonably should have known. Districts must have well publicized policies prohibiting harassment behaviors. If an investigation reveals that a discriminatory harassment occurred, a school must take prompt and effective steps reasonably calculated to end the harassment, eliminate any hostile environment and its effects, and prevent the harassment from recurring.

http://www2.ed.gov/about/offices/list/ocr/letters/colleague-201010.html�

3

A school district, in taking reasonable steps to end harassment, must not penalize the student who was harassed. Remediation duties do not have to be requested by the harassed student. The letter gives certain examples of remedial actions which might be appropriate. Readers are referred to the letter, which is attached, or this link: http://www2.ed.gov/about/offices/list/ocr/letters/colleague-201010.html.

II. Retaliation

Schools must take steps to prevent retaliation against the harassed student who reports the misbehavior or violation. When harassment triggers the civil rights laws, the administration has to do more than just disciplining the perpetrators. The responsibilities include the following:

A. Eliminate the hostile environment created by the harassment;

B. Address the effects of the harassment; and

C. Ensure that the harassment does not recur.

The common approach of having the harassed student confront his or her accusers is seen as an “informal mechanism” and must only be used if it is on a voluntary basis. The practice of simply throwing the kids into the room to talk to each other is seen as largely inappropriate.

http://www2.ed.gov/about/offices/list/ocr/letters/colleague-201010.html�

UNITED STATES DEPARTMENT OF EDUCATION OFFICE FOR CIVIL RIGHTS

October 26, 2010

Dear Colleague:

In recent years, many state departments of education and local school districts have taken steps to reduce bullying in schools. The U.S. Department of Education (Department) fully supports these efforts. Bullying fosters a climate of fear and disrespect that can seriously impair the physical and psychological health of its victims and create conditions that negatively affect learning, thereby undermining the ability of students to achieve their full potential. The movement to adopt anti‐bullying policies reflects schools’ appreciation of their important responsibility to maintain a safe learning environment for all students. I am writing to remind you, however, that some student misconduct that falls under a school’s anti‐bullying policy also may trigger responsibilities under one or more of the federal antidiscrimination laws enforced by the Department’s Office for Civil Rights (OCR). As discussed in more detail below, by limiting its response to a specific application of its anti‐bullying disciplinary policy, a school may fail to properly consider whether the student misconduct also results in discriminatory harassment.

The statutes that OCR enforces include Title VI of the Civil Rights Act of 19641 (Title VI), which prohibits discrimination on the basis of race, color, or national origin; Title IX of the Education Amendments of 19722 (Title IX), which prohibits discrimination on the basis of sex; Section 504 of the Rehabilitation Act of 19733 (Section 504); and Title II of the Americans with Disabilities Act of 19904 (Title II). Section 504 and Title II prohibit discrimination on the basis of disability.5 School districts may violate these civil rights statutes and the Department’s implementing regulations when peer harassment based on race, color, national origin, sex, or disability is sufficiently serious that it creates a hostile environment and such harassment is encouraged, tolerated, not adequately addressed, or ignored by school employees.6 School personnel who understand their legal obligations to address harassment under these laws are in the best position to prevent it from occurring and to respond appropriately when it does. Although this letter focuses on the elementary and secondary school context, the legal principles also apply to postsecondary institutions covered by the laws and regulations enforced by OCR. Some school anti‐bullying policies already may list classes or traits on which bases bullying or harassment is specifically prohibited. Indeed, many schools have adopted anti‐bullying policies that go beyond prohibiting bullying on the basis of traits expressly protected by the federal civil

1 42 U.S.C. § 2000d et seq. 2 20 U.S.C. § 1681 et seq. 3 29 U.S.C. § 794. 4 42 U.S.C. § 12131 et seq. 5 OCR also enforces the Age Discrimination Act of 1975, 42 U.S.C. § 6101 et seq., and the Boy Scouts of America Equal Access Act, 20 U.S.C. § 7905. This letter does not specifically address those statutes. 6 The Department’s regulations implementing these statutes are in 34 C.F.R. parts 100, 104, and 106. Under these federal civil rights laws and regulations, students are protected from harassment by school employees, other students, and third parties. This guidance focuses on peer harassment, and articulates the legal standards that apply in administrative enforcement and in court cases where plaintiffs are seeking injunctive relief.

Our mission is to ensure equal access to education and to promote educational excellence throughout the Nation.

‐ Dear Colleague Letter: Harassment and Bullying

rights laws enforced by OCR—race, color, national origin, sex, and disability—to include such bases as sexual orientation and religion. While this letter concerns your legal obligations under the laws enforced by OCR, other federal, state, and local laws impose additional obligations on schools.7 And, of course, even when bullying or harassment is not a civil rights violation, schools should still seek to prevent it in order to protect students from the physical and emotional harms that it may cause. Harassing conduct may take many forms, including verbal acts and name‐calling; graphic and written statements, which may include use of cell phones or the Internet; or other conduct that may be physically threatening, harmful, or humiliating. Harassment does not have to include intent to harm, be directed at a specific target, or involve repeated incidents. Harassment creates a hostile environment when the conduct is sufficiently severe, pervasive, or persistent so as to interfere with or limit a student’s ability to participate in or benefit from the services, activities, or opportunities offered by a school. When such harassment is based on race, color, national origin, sex, or disability, it violates the civil rights laws that OCR enforces.8 A school is responsible for addressing harassment incidents about which it knows or reasonably

9should have known. In some situations, harassment may be in plain sight, widespread, or well‐known to students and staff, such as harassment occurring in hallways, during academic or physical education classes, during extracurricular activities, at recess, on a school bus, or through graffiti in public areas. In these cases, the obvious signs of the harassment are sufficient to put the school on notice. In other situations, the school may become aware of misconduct, triggering an investigation that could lead to the discovery of additional incidents that, taken together, may constitute a hostile environment. In all cases, schools should have well‐publicized policies prohibiting harassment and procedures for reporting and resolving

10complaints that will alert the school to incidents of harassment. When responding to harassment, a school must take immediate and appropriate action to investigate or otherwise determine what occurred. The specific steps in a school’s investigation will vary depending upon the nature of the allegations, the source of the complaint, the age of the student or students involved, the size and administrative structure of the school, and other factors. In all cases, however, the inquiry should be prompt, thorough, and impartial. If an investigation reveals that discriminatory harassment has occurred, a school must take prompt and effective steps reasonably calculated to end the harassment, eliminate any hostile

7 For instance, the U.S. Department of Justice (DOJ) has jurisdiction over Title IV of the Civil Rights Act of 1964, 42 U.S.C. § 2000c (Title IV), which prohibits discrimination on the basis of race, color, sex, religion, or national origin by public elementary and secondary schools and public institutions of higher learning. State laws also provide additional civil rights protections, so districts should review these statutes to determine what protections they afford (e.g., some state laws specifically prohibit discrimination on the basis of sexual orientation). 8 Some conduct alleged to be harassment may implicate the First Amendment rights to free speech or expression. For more information on the First Amendment’s application to harassment, see the discussions in OCR’s Dear Colleague Letter: First Amendment (July 28, 2003), available at http://www.ed.gov/about/offices/list/ocr/firstamend.html, and OCR’s Revised Sexual Harassment Guidance: Harassment of Students by School Employees, Other Students, or Third Parties (Jan. 19, 2001) (Sexual Harassment Guidance), available at http://www.ed.gov/about/offices/list/ocr/docs/shguide.html. 9 A school has notice of harassment if a responsible employee knew, or in the exercise of reasonable care should have known, about the harassment. For a discussion of what a “responsible employee” is, see OCR’s Sexual Harassment Guidance. 10 Districts must adopt and publish grievance procedures providing for prompt and equitable resolution of student and employee sex and disability discrimination complaints, and must notify students, parents, employees, applicants, and other interested parties that the district does not discriminate on the basis of sex or disability. See 28 C.F.R. § 35.106; 28 C.F.R. § 35.107(b); 34 C.F.R. § 104.7(b); 34 C.F.R. § 104.8; 34 C.F.R. § 106.8(b); 34 C.F.R. § 106.9.

http://www.ed.gov/about/offices/list/ocr/firstamend.html

http://www.ed.gov/about/offices/list/ocr/docs/shguide.html


environment and its effects, and prevent the harassment from recurring. These duties are a school’s responsibility even if the misconduct also is covered by an anti‐bullying policy, and regardless of whether a student has complained, asked the school to take action, or identified the harassment as a form of discrimination. Appropriate steps to end harassment may include separating the accused harasser and the target, providing counseling for the target and/or harasser, or taking disciplinary action against the harasser. These steps should not penalize the student who was harassed. For example, any separation of the target from an alleged harasser should be designed to minimize the burden on the target’s educational program (e.g., not requiring the target to change his or her class schedule). In addition, depending on the extent of the harassment, the school may need to provide training or other interventions not only for the perpetrators, but also for the larger school community, to ensure that all students, their families, and school staff can recognize harassment if it recurs and know how to respond. A school also may be required to provide additional services to the student who was harassed in order to address the effects of the harassment, particularly if the school initially delays in responding or responds inappropriately or inadequately to information about harassment. An effective response also may need to include the issuance of new policies against harassment and new procedures by which students, parents, and employees may report allegations of harassment (or wide dissemination of existing policies and procedures), as well as wide distribution of the contact information for the district’s Title IX and Section 504/Title II coordinators.11 Finally, a school should take steps to stop further harassment and prevent any retaliation against the person who made the complaint (or was the subject of the harassment) or against those who provided information as witnesses. At a minimum, the school’s responsibilities include making sure that the harassed students and their families know how to report any subsequent problems, conducting follow‐up inquiries to see if there have been any new incidents or any instances of retaliation, and responding promptly and appropriately to address continuing or new problems. When responding to incidents of misconduct, schools should keep in mind the following:

• The label used to describe an incident (e.g., bullying, hazing, teasing) does not determine how a school is obligated to respond. Rather, the nature of the conduct itself must be assessed for civil rights implications. So, for example, if the abusive behavior is on the basis of race, color, national origin, sex, or disability, and creates a hostile environment, a school is obligated to respond in accordance with the applicable federal civil rights statutes and regulations enforced by OCR.

• When the behavior implicates the civil rights laws, school administrators should look beyond simply disciplining the perpetrators. While disciplining the perpetrators is likely a necessary step, it often is insufficient. A school’s responsibility is to eliminate the

11 Districts must designate persons responsible for coordinating compliance with Title IX, Section 504, and Title II, including the investigation of any complaints of sexual, gender‐based, or disability harassment. See 28 C.F.R. § 35.107(a); 34 C.F.R. § 104.7(a); 34 C.F.R. § 106.8(a).


hostile environment created by the harassment, address its effects, and take steps to ensure that harassment does not recur. Put differently, the unique effects of discriminatory harassment may demand a different response than would other types of bullying.

Below, I provide hypothetical examples of how a school’s failure to recognize student misconduct as discriminatory harassment violates students’ civil rights.12 In each of the examples, the school was on notice of the harassment because either the school or a responsible employee knew or should have known of misconduct that constituted harassment. The examples describe how the school should have responded in each circumstance. Title VI: Race, Color, or National Origin Harassment

• Some students anonymously inserted offensive notes into African‐American students’

lockers and notebooks, used racial slurs, and threatened African‐American students who tried to sit near them in the cafeteria. Some African‐American students told school officials that they did not feel safe at school. The school investigated and responded to individual instances of misconduct by assigning detention to the few student perpetrators it could identify. However, racial tensions in the school continued to escalate to the point that several fights broke out between the school’s racial groups. In this example, school officials failed to acknowledge the pattern of harassment as indicative of a racially hostile environment in violation of Title VI. Misconduct need not be directed at a particular student to constitute discriminatory harassment and foster a racially hostile environment. Here, the harassing conduct included overtly racist behavior (e.g., racial slurs) and also targeted students on the basis of their race (e.g., notes directed at African‐American students). The nature of the harassment, the number of incidents, and the students’ safety concerns demonstrate that there was a racially hostile environment that interfered with the students’ ability to participate in the school’s education programs and activities. Had the school recognized that a racially hostile environment had been created, it would have realized that it needed to do more than just discipline the few individuals whom it could identify as having been involved. By failing to acknowledge the racially hostile environment, the school failed to meet its obligation to implement a more systemic response to address the unique effect that the misconduct had on the school climate. A more effective response would have included, in addition to punishing the perpetrators, such steps as reaffirming the school’s policy against discrimination (including racial harassment), publicizing the means to report allegations of racial harassment, training faculty on constructive responses to racial conflict, hosting class discussions about racial harassment and sensitivity to students of other races, and conducting outreach to involve parents and students in an effort to identify problems and improve the school climate. Finally, had school officials responded appropriately

12 Each of these hypothetical examples contains elements taken from actual cases.


and aggressively to the racial harassment when they first became aware of it, the school might have prevented the escalation of violence that occurred.13

• Over the course of a school year, school employees at a junior high school received reports of several incidents of anti‐Semitic conduct at the school. Anti‐Semitic graffiti, including swastikas, was scrawled on the stalls of the school bathroom. When custodians discovered the graffiti and reported it to school administrators, the administrators ordered the graffiti removed but took no further action. At the same school, a teacher caught two ninth‐graders trying to force two seventh‐graders to give them money. The ninth‐graders told the seventh‐graders, “You Jews have all of the money, give us some.” When school administrators investigated the incident, they determined that the seventh‐graders were not actually Jewish. The school suspended the perpetrators for a week because of the serious nature of their misconduct. After that incident, younger Jewish students started avoiding the school library and computer lab because they were located in the corridor housing the lockers of the ninth‐graders. At the same school, a group of eighth‐grade students repeatedly called a Jewish student “Drew the dirty Jew.” The responsible eighth‐graders were reprimanded for teasing the Jewish student.

The school administrators failed to recognize that anti‐Semitic harassment can trigger responsibilities under Title VI. While Title VI does not cover discrimination based solely on religion,14 groups that face discrimination on the basis of actual or perceived shared ancestry or ethnic characteristics may not be denied protection under Title VI on the ground that they also share a common faith. These principles apply not just to Jewish students, but also to students from any discrete religious group that shares, or is perceived to share, ancestry or ethnic characteristics (e.g., Muslims or Sikhs). Thus, harassment against students who are members of any religious group triggers a school’s Title VI responsibilities when the harassment is based on the group’s actual or perceived shared ancestry or ethnic characteristics, rather than solely on its members’ religious practices. A school also has responsibilities under Title VI when its students are harassed based on their actual or perceived citizenship or residency in a country whose residents share a dominant religion or a distinct religious identity.15

In this example, school administrators should have recognized that the harassment was based on the students’ actual or perceived shared ancestry or ethnic identity as Jews (rather than on the students’ religious practices). The school was not relieved of its responsibilities under Title VI because the targets of one of the incidents were not actually Jewish. The harassment was still based on the perceived ancestry or ethnic characteristics of the targeted students. Furthermore, the harassment negatively affected the ability and willingness of Jewish students to participate fully in the school’s

13 More information about the applicable legal standards and OCR’s approach to investigating allegations of harassment on the basis of race, color, or national origin is included in Racial Incidents and Harassment Against Students at Educational Institutions: Investigative Guidance, 59 Fed. Reg. 11,448 (Mar. 10, 1994), available at http://www.ed.gov/about/offices/list/ocr/docs/race394.html. 14 As noted in footnote seven, DOJ has the authority to remedy discrimination based solely on religion under Title IV. 15 More information about the applicable legal standards and OCR’s approach to investigating complaints of discrimination against members of religious groups is included in OCR’s Dear Colleague Letter: Title VI and Title IX Religious Discrimination in Schools and Colleges (Sept. 13, 2004), available at http://www2.ed.gov/about/offices/list/ocr/religious‐rights2004.html.

http://www.ed.gov/about/offices/list/ocr/docs/race394.html

http://www2.ed.gov/about/offices/list/ocr/religious-rights2004.html


education programs and activities (e.g., by causing some Jewish students to avoid the library and computer lab). Therefore, although the discipline that the school imposed on the perpetrators was an important part of the school’s response, discipline alone was likely insufficient to remedy a hostile environment. Similarly, removing the graffiti, while a necessary and important step, did not fully satisfy the school’s responsibilities. As discussed above, misconduct that is not directed at a particular student, like the graffiti in the bathroom, can still constitute discriminatory harassment and foster a hostile environment. Finally, the fact that school officials considered one of the incidents “teasing” is irrelevant for determining whether it contributed to a hostile environment.

Because the school failed to recognize that the incidents created a hostile environment, it addressed each only in isolation, and therefore failed to take prompt and effective steps reasonably calculated to end the harassment and prevent its recurrence. In addition to disciplining the perpetrators, remedial steps could have included counseling the perpetrators about the hurtful effect of their conduct, publicly labeling the incidents as anti‐Semitic, reaffirming the school’s policy against discrimination, and publicizing the means by which students may report harassment. Providing teachers with training to recognize and address anti‐Semitic incidents also would have increased the effectiveness of the school’s response. The school could also have created an age‐appropriate program to educate its students about the history and dangers of anti‐Semitism, and could have conducted outreach to involve parents and community groups in preventing future anti‐Semitic harassment.

Title IX: Sexual Harassment

• Shortly after enrolling at a new high school, a female student had a brief romance with another student. After the couple broke up, other male and female students began routinely calling the new student sexually charged names, spreading rumors about her sexual behavior, and sending her threatening text messages and e‐mails. One of the student’s teachers and an athletic coach witnessed the name calling and heard the rumors, but identified it as “hazing” that new students often experience. They also noticed the new student’s anxiety and declining class participation. The school attempted to resolve the situation by requiring the student to work the problem out directly with her harassers. Sexual harassment is unwelcome conduct of a sexual nature, which can include unwelcome sexual advances, requests for sexual favors, or other verbal, nonverbal, or physical conduct of a sexual nature. Thus, sexual harassment prohibited by Title IX can include conduct such as touching of a sexual nature; making sexual comments, jokes, or gestures; writing graffiti or displaying or distributing sexually explicit drawings, pictures, or written materials; calling students sexually charged names; spreading sexual rumors; rating students on sexual activity or performance; or circulating, showing, or creating e‐mails or Web sites of a sexual nature.


In this example, the school employees failed to recognize that the “hazing” constituted sexual harassment. The school did not comply with its Title IX obligations when it failed to investigate or remedy the sexual harassment. The conduct was clearly unwelcome, sexual (e.g., sexual rumors and name calling), and sufficiently serious that it limited the student’s ability to participate in and benefit from the school’s education program (e.g., anxiety and declining class participation). The school should have trained its employees on the type of misconduct that constitutes sexual harassment. The school also should have made clear to its employees that they could not require the student to confront her harassers. Schools may use informal mechanisms for addressing harassment, but only if the parties agree to do so on a voluntary basis. Had the school addressed the harassment consistent with Title IX, the school would have, for example, conducted a thorough investigation and taken interim measures to separate the student from the accused harassers. An effective response also might have included training students and employees on the school’s policies related to harassment, instituting new procedures by which employees should report allegations of harassment, and more widely distributing the contact information for the district’s Title IX coordinator. The school also might have offered the targeted student tutoring, other academic assistance, or counseling as necessary to remedy the effects of the harassment.16

Title IX: Gender‐Based Harassment

• Over the course of a school year, a gay high school student was called names (including anti‐gay slurs and sexual comments) both to his face and on social networking sites, physically assaulted, threatened, and ridiculed because he did not conform to stereotypical notions of how teenage boys are expected to act and appear (e.g., effeminate mannerisms, nontraditional choice of extracurricular activities, apparel, and personal grooming choices). As a result, the student dropped out of the drama club to avoid further harassment. Based on the student’s self‐identification as gay and the homophobic nature of some of the harassment, the school did not recognize that the misconduct included discrimination covered by Title IX. The school responded to complaints from the student by reprimanding the perpetrators consistent with its anti‐bullying policy. The reprimands of the identified perpetrators stopped the harassment by those individuals. It did not, however, stop others from undertaking similar harassment of the student. As noted in the example, the school failed to recognize the pattern of misconduct as a form of sex discrimination under Title IX. Title IX prohibits harassment of both male and female students regardless of the sex of the harasser—i.e., even if the harasser and target are members of the same sex. It also prohibits gender‐based harassment, which may include acts of verbal, nonverbal, or physical aggression, intimidation, or hostility based on sex or sex‐stereotyping. Thus, it can be sex discrimination if students are harassed either for exhibiting what is perceived as a stereotypical characteristic for their

16 More information about the applicable legal standards and OCR’s approach to investigating allegations of sexual harassment is included in OCR’s Sexual Harassment Guidance, available at http://www.ed.gov/about/offices/list/ocr/docs/shguide.html.



sex, or for failing to conform to stereotypical notions of masculinity and femininity. Title IX also prohibits sexual harassment and gender‐based harassment of all students, regardless of the actual or perceived sexual orientation or gender identity of the harasser or target. Although Title IX does not prohibit discrimination based solely on sexual orientation, Title IX does protect all students, including lesbian, gay, bisexual, and transgender (LGBT) students, from sex discrimination. When students are subjected to harassment on the basis of their LGBT status, they may also, as this example illustrates, be subjected to forms of sex discrimination prohibited under Title IX. The fact that the harassment includes anti‐LGBT comments or is partly based on the target’s actual or perceived sexual orientation does not relieve a school of its obligation under Title IX to investigate and remedy overlapping sexual harassment or gender‐based harassment. In this example, the harassing conduct was based in part on the student’s failure to act as some of his peers believed a boy should act. The harassment created a hostile environment that limited the student’s ability to participate in the school’s education program (e.g., access to the drama club). Finally, even though the student did not identify the harassment as sex discrimination, the school should have recognized that the student had been subjected to gender‐based harassment covered by Title IX. In this example, the school had an obligation to take immediate and effective action to eliminate the hostile environment. By responding to individual incidents of misconduct on an ad hoc basis only, the school failed to confront and prevent a hostile environment from continuing. Had the school recognized the conduct as a form of sex discrimination, it could have employed the full range of sanctions (including progressive discipline) and remedies designed to eliminate the hostile environment. For example, this approach would have included a more comprehensive response to the situation that involved notice to the student’s teachers so that they could ensure the student was not subjected to any further harassment, more aggressive monitoring by staff of the places where harassment occurred, increased training on the scope of the school’s harassment and discrimination policies, notice to the target and harassers of available counseling services and resources, and educating the entire school community on civil rights and expectations of tolerance, specifically as they apply to gender stereotypes. The school also should have taken steps to clearly communicate the message that the school does not tolerate harassment and will be responsive to any information about such conduct.17

Section 504 and Title II: Disability Harassment

• Several classmates repeatedly called a student with a learning disability “stupid,” “idiot,” and “retard” while in school and on the school bus. On one occasion, these students tackled him, hit him with a school binder, and threw his personal items into the garbage. The student complained to his teachers and guidance counselor that he was continually being taunted and teased. School officials offered him counseling services and a

17 Guidance on gender‐based harassment is also included in OCR’s Sexual Harassment Guidance, available at http://www.ed.gov/about/offices/list/ocr/docs/shguide.html.



psychiatric evaluation, but did not discipline the offending students. As a result, the harassment continued. The student, who had been performing well academically, became angry, frustrated, and depressed, and often refused to go to school to avoid the harassment. In this example, the school failed to recognize the misconduct as disability harassment under Section 504 and Title II. The harassing conduct included behavior based on the student’s disability, and limited the student’s ability to benefit fully from the school’s education program (e.g., absenteeism). In failing to investigate and remedy the misconduct, the school did not comply with its obligations under Section 504 and Title II. Counseling may be a helpful component of a remedy for harassment. In this example, however, since the school failed to recognize the behavior as disability harassment, the school did not adopt a comprehensive approach to eliminating the hostile environment. Such steps should have at least included disciplinary action against the harassers, consultation with the district’s Section 504/Title II coordinator to ensure a comprehensive and effective response, special training for staff on recognizing and effectively responding to harassment of students with disabilities, and monitoring to ensure that the harassment did not resume.18

I encourage you to reevaluate the policies and practices your school uses to address bullying19 and harassment to ensure that they comply with the mandates of the federal civil rights laws. For your convenience, the following is a list of online resources that further discuss the obligations of districts to respond to harassment prohibited under the federal antidiscrimination laws enforced by OCR:

• Sexual Harassment: It’s Not Academic (Revised 2008): http://www.ed.gov/about/offices/list/ocr/docs/ocrshpam.html

• Dear Colleague Letter: Sexual Harassment Issues (2006):

http://www2.ed.gov/about/offices/list/ocr/letters/sexhar‐2006.html

• Dear Colleague Letter: Religious Discrimination (2004): http://www2.ed.gov/about/offices/list/ocr/religious‐rights2004.html

• Dear Colleague Letter: First Amendment (2003): http://www.ed.gov/about/offices/list/ocr/firstamend.html

18 More information about the applicable legal standards and OCR’s approach to investigating allegations of disability harassment is included in OCR’s Dear Colleague Letter: Prohibited Disability Harassment (July 25, 2000), available at http://www2.ed.gov/about/offices/list/ocr/docs/disabharassltr.html. 19 For resources on preventing and addressing bullying, please visit http://www.bullyinginfo.org, a Web site established by a federal Interagency Working Group on Youth Programs. For information on the Department’s bullying prevention resources, please visit the Office of Safe and Drug‐Free Schools’ Web site at http://www.ed.gov/offices/OESE/SDFS. For information on regional Equity Assistance Centers that assist schools in developing and implementing policies and practices to address issues regarding race, sex, or national origin discrimination, please visit http://www.ed.gov/programs/equitycenters.

http://www.ed.gov/about/offices/list/ocr/docs/ocrshpam.html

http://www2.ed.gov/about/offices/list/ocr/letters/sexhar-2006.html

http://www2.ed.gov/about/offices/list/ocr/religious-rights2004.html

http://www.ed.gov/about/offices/list/ocr/firstamend.html

http://www2.ed.gov/about/offices/list/ocr/docs/disabharassltr.html

http://www.bullyinginfo.org/

http://www.ed.gov/offices/OESE/SDFS

http://www.ed.gov/programs/equitycenters


• Sexual Harassment Guidance (Revised 2001): http://www.ed.gov/about/offices/list/ocr/docs/shguide.html

• Dear Colleague Letter: Prohibited Disability Harassment (2000): http://www.ed.gov/about/offices/list/ocr/docs/disabharassltr.html

• Racial Incidents and Harassment Against Students (1994): http://www.ed.gov/about/offices/list/ocr/docs/race394.html

Please also note that OCR has added new data items to be collected through its Civil Rights Data Collection (CRDC), which surveys school districts in a variety of areas related to civil rights in education. The CRDC now requires districts to collect and report information on allegations of harassment, policies regarding harassment, and discipline imposed for harassment. In 2009‐10, the CRDC covered nearly 7,000 school districts, including all districts with more than 3,000 students. For more information about the CRDC data items, please visit http://www2.ed.gov/about/offices/list/ocr/whatsnew.html. OCR is committed to working with schools, students, students’ families, community and advocacy organizations, and other interested parties to ensure that students are not subjected to harassment. Please do not hesitate to contact OCR if we can provide assistance in your efforts to address harassment or if you have other civil rights concerns. For the OCR regional office serving your state, please visit: http://wdcrobcolp01.ed.gov/CFAPPS/OCR/contactus.cfm, or call OCR’s Customer Service Team at 1‐800‐421‐3481. I look forward to continuing our work together to ensure equal access to education, and to promote safe and respectful school climates for America’s students.

Sincerely, /s/

Russlynn Ali Assistant Secretary for Civil Rights


http://www.ed.gov/about/offices/list/ocr/docs/disabharassltr.html

http://www.ed.gov/about/offices/list/ocr/docs/race394.html

http://www2.ed.gov/about/offices/list/ocr/whatsnew.html

http://wdcrobcolp01.ed.gov/CFAPPS/OCR/contactus.cfm

1



Ph. 847-564-8662 Fax 847-564-8419



Notes #5



2

Bullying Notes Continued

These are simply notes taken by Mr. Whitted from reading provided by the Task Force Coordinator, Peggie Garcia. Document reviewed includes the 07/15/09 Education Week. Relevant points: 1. Bullying needs to be addressed with the same gravity with which we address other problems disrupting learning. The message should be that it will not be tolerated.

BRW note: I have always said that the only “zero tolerance” should be zero tolerance for bullying and harassment.

2. The Education Week article makes the very good point that students are asked to do math and other problems repeatedly; the same should be true for issues involving bullying and harassment; the lessons should be given frequently. 3. Schools must also have in place a reliable system for letting parents know when and where harassment is occurring. It was pointed out in the article that the Education Department were let out grants for school climate surveys, then there will be safety scores by school. School leaders must create a climate where students know there are adults they can trust and to whom they can safely report information. In my experience, this is the weak spot in most districts. 4. Safe reporting can involve anonymous tip lines; manuals that require staff to report acts of bullying and keep victims safe from retaliation; and a focus on bystanders who can protect victims, as in Finland’s KiVa system. 5. Less than half of principals surveyed feel that the children of other than traditional sexual persuasion feel safe at their school. Arnie Duncan has appointed Mr. Kevin Jennings to head the Safe and Drug Free Schools Department at the U.S. Department of Education. Sexual orientation is the second most common reason for school bullying.

http://www.edweek.org/ew/articles/2009/06/22/36jennings.h28.html?r=1308221761�

3

6. When teachers avoid discussing homophobia, kids get the message that homophobic epithets are acceptable. Teaching tolerance should be as much a part of the curriculum as Algebra. 7. BRW note: Although it is true that only two groups of individuals are restricted to buildings (prisoners and students) during certain times, schools should not encourage a prison yard mentality. 8. BRW note: How about looking at bullying as an expression of violence, not just of individuals, but also of systems of individuals caught in an institutional order that can be psychological alienating, disrespectful and oppressive? 9. The “shadow side of school,” created by the very nature of our educational system, engenders a culture of hostility, fear, shame, excessive competition, and lack of respect for differences. 10. Consistent (and fair) consequences, used well, undermine traditional destructive privilege and protect those who are less powerful.

|88| Anti harassment bullying task force materials (combined)

Education

Transcript of |88| Anti harassment bullying task force materials (combined)