Abstract
Objectives. To compare changes over time in prevalence of school victimization among lesbian, gay, and bisexual (LGB) students compared with heterosexual students.
Methods. We analyzed data from 11 Youth Risk Behavior Surveys conducted among representative samples of students in grades 9 through 12 in Massachusetts during 1995 to 2015. We used multivariable logistic regression models to identify trends over time by sexual identity.
Results. During 1995 to 2015, the prevalence of missing school decreased overall (from 5.6% to 4.8%) and among heterosexual (from 4.3% to 3.8%) and LGB (from 25.0% to 13.4%) students. The prevalence of having been threatened decreased overall (from 7.8% to 4.1%) and among heterosexual (from 6.5% to 3.5%) and LGB (from 32.9% to 6.7%) students.
Conclusions. We identified evidence of a significant decrease in victimization among all students regardless of sexual identity and a steep decline among LGB students. Additional actions to improve school climate may help eliminate the disparities and decrease victimization for all youths.
Lesbian, gay, and bisexual (LGB) youths routinely experience violence and bullying more frequently than do their heterosexual peers; this disparity is well-documented in the literature.1,2 However, no studies have assessed long-term trends in victimization among LGB youths compared with heterosexual youths.
We compared changes over time in prevalence of school-related victimization by sexual identity by using Massachusetts Youth Risk Behavior Survey (MYRBS) data from 1995 to 2015. Based on the increased acceptance of LGB individuals3 and improvements to school climate in recent years,4 we hypothesized that school victimization has decreased during 1995 to 2015 for all sexual identity subgroups, but that the magnitude of the trend (i.e., the slope) may differ for heterosexual compared with LGB students.
METHODS
The MYRBS has been conducted biennially since 1991 among population-based representative samples of public school students in grades 9 through 12 in the state of Massachusetts. Participating students completed an anonymous, voluntary questionnaire during a single class period. More information about the survey methods, which are consistent between cycles, are published elsewhere.5
In 1995, 2 questions were added to the MYRBS to ascertain sexual orientation, including 1 question assessing sexual identity. The MYRBS was one of the first population-based surveys of youths to ascertain sexual orientation. Sexual identity was ascertained with the question, “Which of the following best describes you?” and response options, “Heterosexual or straight,” “gay or lesbian,” “bisexual,” and “not sure.” For this analysis, “gay or lesbian” was combined with “bisexual” to create a LGB category. Students who selected “not sure” remained in the models but their results are not reported separately. The 1995 to 2015 MYRBSs also assessed having not gone to school because of safety concerns and having been threatened or injured with a weapon on school property; in addition, in 2009, a question was added to assess bullying victimization on school property.
With SUDAAN version 11 (Research Triangle Institute, Research Triangle Park, NC) to account for the complex sample design and sampling weights, we used the t test to assess differences in unadjusted prevalence estimates. The trend analysis used logistic regression to separately model secular trends in each school violence measure; all models set sexual identity as the exposure variable (with heterosexual students as the reference group) and adjusted for sex and race/ethnicity. We coded continuous linear and quadratic time variables by using orthogonal coefficients. Because the bullying measure was only assessed in 4 MYRBS cycles, we did not test for quadratic time components in this model only. Additional information about trend analysis methods can be found in Appendix A (available as a supplement to the online version of this article at http://www.ajph.org).
RESULTS
During 1995 to 2015, the MYRBS was conducted 11 times. The sample sizes ranged from 2707 (in 2009) to 4415 (in 1999; median = 3522); the school response rates ranged from 75% (in 2015) to 96% (in 1999 and 2001; median = 87%); the student response rates ranged from 77% (in 1995) to 88% (in 2013; median = 81%); and the overall response rates ranged from 61% (in 2015) to 77% (in 2001; median = 70%).
In 1995 and 2015, the prevalence of having not gone to school because of safety concerns was higher among LGB than heterosexual students (Table 1). In 2009 and 2015, the prevalence of having been bullied on school property was also higher among LGB than heterosexual students. However, although the prevalence of having been threatened or injured with a weapon on school property was higher among LGB than heterosexual students in 1995, in 2015 we detected no differences in the prevalence of this behavior by sexual identity.
TABLE 1—
Prevalence and Trends of School-Related Victimization Among Students in Grades 9 Through 12, by Sexual Identity: Massachusetts Youth Risk Behavior Surveys, 1995–2015
| Did Not Go to School Because of Safety Concerns,a Prevalence (95% CI) |
Threatened or Injured With a Weapon on School Property,b Prevalence (95% CI) |
Bullied on School Property,c Prevalence (95% CI) |
|||||||
| Year | Overall | Heterosexual | LGB | Overall | Heterosexual | LGB | Overall | Heterosexual | LGB |
| 1995 | 5.6 (4.4, 7.2) | 4.3 (3.4, 5.5) | 25.0 (15.4, 37.8) | 7.8 (6.6, 9.1) | 6.5 (5.5, 7.7) | 32.9 (24.3, 42.7) | . . . | . . . | . . . |
| 1997 | 4.6 (3.9, 5.6) | 3.9 (3.2, 4.8) | 22.2 (13.5, 34.3) | 7.5 (6.6, 8.5) | 6.5 (5.7, 7.3) | 31.2 (20.2, 44.8) | . . . | . . . | . . . |
| 1999 | 6.4 (5.0, 8.0) | 5.3 (4.1, 7.0) | 20.0 (13.0, 29.4) | 8.6 (7.7, 9.6) | 7.6 (6.8, 8.5) | 24.4 (16.2, 35.1) | . . . | . . . | . . . |
| 2001 | 8.1 (6.1, 10.6) | 7.7 (5.7, 10.3) | 17.6 (11.9, 25.4) | 8.2 (7.2, 9.3) | 7.6 (6.7, 8.7) | 17.4 (11.8, 25.0) | . . . | . . . | . . . |
| 2003 | 4.6 (4.0, 5.3) | 3.9 (3.3, 4.5) | 16.3 (10.4, 24.6) | 6.3 (5.3, 7.4) | 5.4 (4.6, 6.3) | 19.3 (13.3, 27.2) | . . . | . . . | . . . |
| 2005 | 4.0 (3.4, 4.7) | 3.3 (2.7, 4.0) | 16.5 (11.4, 23.1) | 5.4 (4.6, 6.4) | 5.0 (4.2, 6.1) | 10.3 (5.5, 18.7) | . . . | . . . | . . . |
| 2007 | 4.7 (3.9, 5.7) | 3.8 (3.0, 4.9) | 13.0 (9.4, 17.7) | 5.3 (4.5, 6.3) | 4.3 (3.5, 5.3) | 18.7 (13.4, 25.5) | . . . | . . . | . . . |
| 2009 | 4.0 (3.1, 5.2) | 3.1 (2.2, 4.2) | 13.9 (8.4, 22.2) | 7.0 (6.0, 8.2) | 6.2 (5.1, 7.5) | 17.3 (12.4, 23.7) | 19.4 (17.7, 21.2) | 17.3 (15.7, 19.0) | 42.3 (35.6, 49.3) |
| 2011 | 4.8 (3.9, 6.1) | 4.2 (3.3, 5.5) | 9.9 (6.2, 15.6) | 6.8 (5.6, 8.2) | 5.6 (4.5, 7.1) | 15.2 (10.5, 21.5) | 18.1 (16.1, 20.2) | 16.4 (14.5, 18.5) | 33.5 (26.7, 41.1) |
| 2013 | 3.6 (2.9, 4.5) | 2.9 (2.2, 3.8) | 15.3 (9.6, 23.4) | 4.3 (3.7, 5.2) | 3.9 (3.2, 4.7) | 14.0 (10.0, 19.3) | 16.6 (14.7, 18.6) | 15.5 (13.6, 17.6) | 37.7 (29.9, 46.2) |
| 2015 | 4.8 (4.0, 5.8) | 3.8 (2.9, 4.8) | 13.4 (8.9, 19.7) | 4.1 (3.2, 5.1) | 3.5 (2.7, 4.5) | 6.7 (3.6, 12.3) | 15.6 (14.0, 17.3) | 13.8 (12.3, 15.4) | 34.3 (27.4, 41.9) |
| Linear Bd (SE) | −0.697 (0.147) | −0.702 (0.166) | −0.798 (0.347) | −0.879 (0.098) | −0.837 (0.104) | −1.275 (0.310) | −1.041 (0.286) | −0.867 (0.294) | −0.902 (0.789) |
| P B | < .001 | < .001 | .022 | < .001 | < .001 | < .001 | < .001 | .003 | .25 |
| Conclusionse | Linear decrease | Linear decrease | Linear decrease | Linear decrease | Linear decrease | Linear decrease | Linear decrease | Linear decrease | No change |
Note. CI = confidence interval; LGB = lesbian, gay, or bisexual. Ellipses indicate that data were not available for that year.
Did not go to school on 1 or more days during the 30 days before the survey because they felt they would be unsafe at school or on the way to or from school.
Were threatened or injured with a weapon (e.g., a gun, knife, or club) on school property 1 or more times during the 12 months before the survey.
Were bullied on school property during the 12 months before the survey.
Parameter estimate for the B associated with the linear time component regressed on the victimization outcome, stratified by sexual identity and controlled for sex and race/ethnicity.
Interpretation of the linear time components.
During 1995 to 2015, when we controlled for race/ethnicity and sex, the prevalence of having not gone to school because of safety concerns decreased linearly overall (from 5.6% to 4.8%), among heterosexual students (from 4.3% to 3.8%), and among LGB students (from 25.0% to 13.4%). The prevalence of having been threatened or injured with a weapon on school property also decreased linearly overall (from 7.8% to 4.1%), among heterosexual students (from 6.5% to 3.5%), and among LGB students (from 32.9% to 6.7%). Furthermore, during 2009 to 2015, when we controlled for race/ethnicity and sex, the prevalence of having been bullied on school property decreased linearly overall (from 19.4% to 15.6%) and among heterosexual students (from 17.3% to 13.8%), but we did not detect a linear change among LGB students. We did not detect quadratic trends overall or by subgroup for any of the 3 behaviors.
DISCUSSION
The results of this analysis support our hypotheses and indicate that school-related violence victimization significantly declined among both heterosexual and LGB students during 1995 to 2015. However, this was not true for being bullied on school property. Although being bullied on school property was not assessed on the MYRBS until 2009, during the past 4 MYRBS cycles, being bullied on school property has declined significantly among heterosexual but not among LGB students. To note, current prevalence of some school victimization is still higher among LGB than heterosexual students.
This analysis is subject to several limitations. The data are only generalizable to Massachusetts students who are enrolled in public school; furthermore, LGB youths might represent a disproportionate percentage of high-school dropouts and other youths who are absent from or do not attend school.6 In addition, some students might not have known their sexual identity; might have been unwilling to disclose it on the MYRBS questionnaire; might have been unwilling to label themselves as heterosexual, gay, lesbian, or bisexual; or might not have understood the sexual identity question. Next, the extent of underreporting or overreporting of health-related behaviors cannot be determined, although the survey questions demonstrate good test–retest reliability.7
Finally, possibly because of the small prevalence of LGB identity, we were unable to detect significant differences in the linear time components between heterosexual and LGB students. Replicating these analyses on larger data sets may yield more convincing results. Despite these limitations, these results are the first to assess secular trends of victimization by sexual identity among large, population-based samples of youths.
PUBLIC HEALTH IMPLICATIONS
Since the mid-1990s, societal acceptance of LGB individuals has increased,3 and school-related victimization of LGB students has decreased. Nonetheless, the prevalence of victimizations remains too high, and it is important to eliminate school-related victimization for all students. Previous studies suggest that victims of school violence and bullying are at greater risk for many other health-risk behaviors,8 and may have worse long-term outcomes compared with their nonvictimized peers,9 especially for LGB youths.10 Schools can take action to improve school climate, which may help eliminate disparities and decrease victimization for all youths.
In 2010, Massachusetts enacted comprehensive legislation to address bullying in public and nonpublic schools and require every school to have a bullying prevention plan (Mass Gen Laws, Ch 71, § 37O [2010, 2014]). The law was amended in 2014 to require schools to “recognize” in their bullying prevention plan that certain enumerated categories of students may be more vulnerable to being bullied based on actual or perceived differentiating characteristics. The Massachusetts Safe Schools Program for lesbian, gay, bisexual, transgender, and queer (LGBTQ) students is designated specifically to help schools implement all state laws that have an impact on LGBTQ students, including this antibullying law. A recent report suggests that Massachusetts schools increased their ability to facilitate access to health services for LGBTQ youths, implementation of “safe spaces,” prohibition of harassment on the basis of sexual orientation, and staff development on safe and supportive school environments.11 These policies and practices aim to reduce victimizations and bullying for all students and particularly for LGB students, and recent research has indicated that they may be effective at preventing and decreasing bullying.12
ACKNOWLEDGMENTS
The authors thank the Massachusetts Department of Elementary and Secondary Education for permission to use the Massachusetts Youth Risk Behavior Survey (MYRBS) data, and they also thank Tim McManus, MS, for his work creating the combined trend data set.
HUMAN PARTICIPANT PROTECTION
The contractor for the MYRBS, the Center for Survey Research at UMASS Boston, determined that the MYRBS is surveillance and not research and therefore it is exempt from full institutional review board review. Survey protocols were submitted to the institutional review board.
REFERENCES
- 1.Kann L, Olsen EO, McManus T et al. Sexual identity, sex of sexual contacts, and health-related behaviors among students in grades 9–12—United States and selected sites, 2015. MMWR Surveill Summ. 2016;65(9):1–202. doi: 10.15585/mmwr.ss6509a1. [DOI] [PubMed] [Google Scholar]
- 2.Kosciw JG, Greytak EA, Palmer NA, Boesen MJ. The 2013 National School Climate Survey: The Experiences of Lesbian, Gay, Bisexual and Transgender Youth in Our Nation’s Schools. Gay, Lesbian, and Straight Education Network. 2014. Available at: https://www.glsen.org/sites/default/files/2013%20National%20School%20Climate%20Survey%20Full%20Report_0.pdf. Accessed March 21, 2017.
- 3.Keleher A, Smith ER. Growing support for gay and lesbian equality since 1990. J Homosex. 2012;59(9):1307–1326. doi: 10.1080/00918369.2012.720540. [DOI] [PubMed] [Google Scholar]
- 4. Centers for Disease Control and Prevention. Results from the School Health Policies and Practices Study 2014. 2015. Available at: http://www.cdc.gov/healthyyouth/data/profiles/pdf/2014/2014_profiles_report.pdf. Accessed October 24, 2016.
- 5.Brener ND, Kann L, Shanklin S et al. Methodology of the Youth Risk Behavior Surveillance System—2013. MMWR Recomm Rep. 2013;62(RR-1):1–20. [PubMed] [Google Scholar]
- 6.Burton CM, Marshal MP, Chisholm DJ. School absenteeism and mental health among sexual minority youth and heterosexual youth. J Sch Psychol. 2014;52(1):37–47. doi: 10.1016/j.jsp.2013.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Brener ND, Kann L, McManus T, Kinchen SA, Sundberg EC, Ross JG. Reliability of the 1999 Youth Risk Behavior Survey questionnaire. J Adolesc Health. 2002;31(4):336–342. doi: 10.1016/s1054-139x(02)00339-7. [DOI] [PubMed] [Google Scholar]
- 8.Hertz MF, Everett Jones S, Barrios L, David-Ferdon C, Holt M. Association between bullying victimization and health risk behaviors among high school students in the United States. J Sch Health. 2015;85(12):833–842. doi: 10.1111/josh.12339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Copeland WE, Wolke D, Angold A, Costello EJ. Adult psychiatric outcomes of bullying and being bullied by peers in childhood and adolescence. JAMA Psychiatry. 2013;70(4):419–426. doi: 10.1001/jamapsychiatry.2013.504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rivers I. The bullying of sexual minorities at school: its nature and long-term correlates. Educ Child Psychol. 2001;18(1):32–46. [Google Scholar]
- 11. 2014 School Health Profiles Report: Trend Analysis Report—Principal Survey. Malden, MA: Massachusetts Department of Elementary and Secondary Education; 2015.
- 12.Hatzenbuehler ML, Schwab-Reese L, Ranapuwala SI, Hertz MF, Ramirez MR. Associations between antibullying policies and bullying in 25 states. JAMA Pediatr. 2015;169(10):e152411. doi: 10.1001/jamapediatrics.2015.2411. [DOI] [PMC free article] [PubMed] [Google Scholar]