Preventing and Responding to Teen Dating Violence: A National Study of School Principals' Perspectives and Practices

Abstract

Teen dating violence (TDV) has emerged as a significant child and adolescent social and health problem. Schools are being viewed as an avenue for TDV prevention. Knowledge about school administrators' perspectives and practices on preventing and responding to TDV can help inform better practice and policymaking on TDV. A national random sample (n = 750) of high school principals in the United States were sent a valid and reliable questionnaire on TDV prevention practices (response rate = 54%). Majority of the school principals reported that: they never received formal training on TDV (68%), their school did not have a protocol to respond to an incident of TDV (76%), training to assist TDV victims was not provided to personnel in their schools in the past 2 years (62%), and the school violence prevention policy did not specifically address TDV (65%). Majority of the school principals had assisted a TDV victim in the past 2 years (57%), but most did not sanction disciplinary actions for TDV perpetrators (73%). Most of the responding principals could not answer four out of nine questions on a TDV knowledge scale. The most common ways to assist victims of TDV by school principals were: referral to school counselor (93%), informing parents or guardians (85%), and informing police or legal authorities (74%). Majority of the school principals did not believe that health teachers (53%) or peers (62%) could play a major role in assisting victims of TDV or preventing TDV. Selected predictors of whether or not a school assisted TDV victims were identified in regression analyses (e.g., training of school personnel and various school policies). School principals identified a variety of barriers to assisting TDV victims, including lack of training and appropriate protocols to respond to TDV incidents. Based on this national assessment of school principals' perspectives, current practices in schools call for greater emphasis on TDV prevention. Implications for TDV prevention practice and policy formulation are discussed based on study findings.

Introduction

Teen dating violence (TDV) has emerged as a significant child and adolescent health and social problem. According to various estimates, 5–35% of teenagers in the United States report being in an abusive romantic relationship every year (Glass et al. 2003; Haynie et al. 2013; Howard et al. 2013). In a recent study, based on the National Youth Risk Behavior Survey, it was reported that from 1999 to 2011, almost a tenth of all high school students experienced physical dating violence every year (Rothman and Xuan 2014). Another report from the same survey found that more than a fifth (20.9%) of adolescent females and more than a tenth (10.4%) of adolescent males were victimized in a dating relationship in 2013 (including physical and sexual abuse) (Vagi et al. 2015). TDV can have minor to long-lasting physical and mental health effects on adolescents. Minor problems include injuries, bruises, and verbal or emotional abuse; major types include rape or even murder. Even if TDV is minor, severe consequences such as depression or suicidal ideation can emerge in victims. Also, victimization in an adolescent romantic relationship confers distinct risk profiles on youth such as engagement in other violent behaviors, psychosocial distress, sexually transmitted diseases, teenage pregnancy, and drug use, just to name a few (Howard et al. 2007, 2008, 2013; Vagi et al. 2015).

The attention given to TDV over the past decade has spurred individual states to take a step forward and enact laws aiming at preventing TDV through school-based interventions, with more than 20 states having a TDV prevention law by 2015 (National Conference of State Legislatures 2017). The major components of these laws include the requirements for schools to teach students about healthy relationships, dating violence prevention, and educating students on reporting procedures for incidents dating violence, in addition to providing staff and personnel training on TDV prevention (Hoefer et al. 2015; National Conference of State Legislatures 2017). As schools are increasingly becoming an avenue of preventive and remedial measures for TDV, school personnel are being viewed as important change agents for preventing and responding to TDV incidents (Fellmeth et al. 2013; Hoefer et al. 2015; Khubchandani et al. 2013). Recently, two studies of high school nurses and counselors found that the majority of respondents had assisted student victims of TDV within the past 2 years. However, the vast majority of school nurses (88%) and counselors (90%) also reported that formal training to assist TDV victims was not provided by their schools. School nurses and counselors reported barriers to effective TDV prevention and response such as lack of trained personnel, lack of parental approval, and lack of support from school administration (Khubchandani et al. 2012, 2013). School administrators have a major role to play in preventing and responding to youth violence (CDC 2016; Dela-Rue et al. 2017; Fellmeth et al. 2013). However, little is known about school administrators' practices in relation to TDV prevention in U.S. schools. Thus, the purpose of this study was to conduct a national assessment of school principals' perspectives and practices on preventing and responding to TDV incidents.

Materials and Methods

Subjects and procedures

A national random sample of high school principals was selected from the database of 20,700 U.S. high schools listed on the National Center for Education Statistics website. An a priori power analysis was completed with a conservative 50/50 split with regard to the practice of interest, and it was determined that a sample of 378 principals would be needed to make inferences to the total population of 20,700 high school principals in the United States with a sampling error of ±5% at the 95% confidence level (Price et al. 2016; U.S. Department of Education 2016). Factoring in a potential nonresponse rate of ∼50%, 750 principals were randomly selected to receive surveys. Several techniques published in the literature on school professionals' research participation were used to help maximize the survey response rate (Dake et al. 2004; Price et al. 2016; Rickard et al. 2010). The study design and protocol were approved by the Institutional Review Board for Human Subjects' Research before the mailings were initiated.

Instrument

A four page, 33-item (multi-component) instrument was developed to assess high school principals' perspectives and practices regarding TDV prevention. To ensure face validity, the questionnaire was based on a comprehensive review of literature and survey items from other studies with school personnel regarding violence prevention in schools (Dake et al. 2004; Khubchandani et al. 2013; Price et al. 2016; Vagi et al. 2013). To establish content validity, the instrument was mailed for review to a panel of published experts (n = 9) in the areas of school health, pediatrics, and survey research as well as school personnel (n = 3). After the expert review, changes were made to the instrument (wording changes and deleting or adding items) to ensure that valid measures regarding perceptions and practices of high school principals regarding TDV were included.

The first set of questions asked the principals to report their schools' practices and policies regarding TDV prevention (i.e., whether students were being educated, whether school personnel provided formal training on TDV prevention, whether the school had a policy on TDV prevention, or whether their schools conduct student behavior assessments to evaluate the prevalence of TDV). A series of items (n = 9) were used as a scale to assess the knowledge of school principals on TDV-related issues. Another series of items assessing school principals' perspectives on the role of various school personnel in preventing and responding to TDV incidents formed a separate scale.

The instrument also assessed the perceived barriers to assisting victims of TDV and whether the respondent's school had a TDV incident response protocol. The Stages of Change component of the Trans-Theoretical Model was used to assess whether high school principals' schools had a protocol to deal with incidents of TDV (Prochaska et al. 2001). This concept assesses the progress of groups as they move from not thinking about a behavior (precontemplation stage) to having been actively involved in a behavior for longer than a year (maintenance stage). The Stages of Change component is mostly used with personal behaviors but has also been used for assessing organizational behavior changes (Prochaska et al. 2001; Rickard et al. 2010). In addition, the instrument also included questions about high school principals' perceptions of the extent of the TDV problem in U.S. schools and their own schools, whether school principals assisted victims of TDV, sanctioned disciplinary actions against TDV perpetrators in the past 2 years, and the perceived ways to assist TDV victims. Demographic and background characteristics of respondents were assessed.

Internal consistency reliabilities were established for five subscales by using the final responses to the instrument. The five subscales and their internal reliabilities were: school practices and policies (α = 0.70), school principals' knowledge about TDV (α = 0.72), school principals' perceptions about ways to assist TDV victims (α = 0.81), perceived roles of various individuals in assisting TDV victims (α = 0.63), and beliefs about the role of school personnel in preventing TDV (α = 0.90).

Data analysis

Data from the study were analyzed by using SPSS 23.0. Data analysis included descriptive statistics for questionnaire items (i.e., frequencies and means). T-tests were calculated to determine differences between dichotomous independent and continuous dependent variables. Chi-Square tests were utilized to assess group differences across categorical variables. Logistic regression analyses with adjusted odds ratios (AORs) were conducted to determine predictors of an outcome variable (assistance provided to TDV victims) based on independent predictor variables (e.g., school and principal characteristics).

Results

Principals' demographic characteristics and TDV-related knowledge

Out of the 750 copies of the instrument that were mailed, 18 copies could not be used (incorrect address, undeliverable, etc.). A total of 396 out of 732 principals responded (54%). The vast majority of the respondents were: males (69.2%), White (86.6%), 40–59 years old (72.1%), and held certification or held principal licensure. Less than one-third (32.3%) of the principals had any formal training on TDV (Table 1).

Table 1.

Demographic and Background Characteristics of School Principals

Variable	Category	N (%)
Sex	Female	122 (30.8)
Sex	Male	274 (69.2)
Age	20–29 years	4 (1)
	30–39 years	60 (15.2)
	40–49 years	149 (37.7)
	50–59 years	131 (33)
	60+ years	47 (11.8)
Race/ethnicity	African American	30 (7.7)
	Asian	1 (.3)
	Hispanic	15 (3.7)
	White	337 (85.2)
	Other	8 (2.0)
Location of school	Urban	90 (22.7)
	Suburban	115 (29)
	Rural	190 (48)
Highest level of education	Bachelors	4 (1.0)
	Masters	237 (59.9)
	Specialist	110 (27.9)
	Doctorate	41 (10.5)
Employment status	Full time	392 (99)
Employment status	Part time	4 (1)
Certification status	Yes	369 (93.3)
Certification status	No	20 (5)
Administrators license	Yes	356 (89.9)
Administrators license	No	11 (2.7)
Received formal training on TDV	Yes	128 (32.3)
Received formal training on TDV	No	268 (67.7)

		Mean ± (SE)
Years worked as school principal	Range (0–40 years)	8.13 (0.44)
Number of students in school	Range (45–3400 students)	940.80 (44.22)
Student population	% Whites	71.23 (1.72)
Student population	% Non-Whites	28.68 (1.69)

N = 396, the numbers may not add up to 100% due to missing values.

TDV, teen dating violence; SE, standard error.

Respondents were asked to rate their perceptions of the extent of the TDV problem in U.S. schools and in their own schools on a scale of 1 (no problem) to 5 (major problem). The average perceived extent of the TDV problem in U.S. schools was 3.10 (standard error [SE] = ± 0.06), and the average extent of the TDV problem in the principals' school was 2.27 (SE = ± 0.04). The median perception of the extent of TDV for U.S. schools was three and that of the school of the respondents was two. The perceived extent of the TDV problem in the principals' own schools was directly correlated with: the number of TDV victims assisted by the principals in the past 2 years (r = 0.63, p = 0.01), and the number of perpetrators who received a TDV-related disciplinary action in the past 2 years from the principals (r = 0.45, p = 0.03).

A series of factual statements were used to assess the knowledge of high school principals' knowledge regarding TDV. Of the nine questions, five were answered correctly by more than half of the principals (Table 2). A knowledge score was computed for the entire population of respondents. For each question answered correctly, the respondent received a score of 1, and for each question answered incorrectly, the respondent received a score of 0. However, the possible range for knowledge scores was 0–9; the maximum score obtained was 8. The average score for the population was 4.87 (SE = ± 0.07), and the median score was 5. The knowledge score correlated directly with the perceived extent of dating violence in the respondent's school (r = 0.54, p = 0.02) and the number of TDV victims assisted in the past 2 years (r = 0.69, p = 0.009).

Table 2.

School Principals' Knowledge on Teen Dating Violence

Item (correct answer)	Answered correctly N (%)
Patterns of dating violence behavior often start early in adolescence and carry through into adult relationships. (True)	368 (93)
Children who have been abused are more likely to become perpetrators or victims of dating violence. (True)	360 (91)
Dating abuse can lead to risky sexual behaviors that can result in unintended pregnancy, sexually transmitted diseases, and HIV infections. (True)	341 (86)
Less than 5% of high school students experience physical dating violence. (False)	210 (53)
Girls who report physical or sexual dating abuse have higher rates of drug, alcohol, and tobacco use than girls who report no abuse. (True)	198 (50)
Victims of dating violence typically talk about the abuse with their peers. (True)	123 (31)
Abuse in a dating relationship occurs more commonly in students from lower socioeconomic backgrounds compared with students from higher socioeconomic backgrounds. (True)	111 (28)
Adolescent dating violence occurs more frequently among racial and ethnic minorities as compared with Whites. (True)	63 (16)
Physical dating violence is more common against adolescent females than males. (False)	51 (13)

N = 396.

School policies and practices regarding TDV

Based on the Stages of Change theory, the school principals were asked to identify the stage they were in, as it related to having a protocol for responding to an incident of TDV. The majority (76%) of the school principals reported that they did not have a school protocol or procedure to respond to an incident of TDV [precontemplation: “we have never seriously thought about creating a protocol for responding to a dating violence incident” (51%); contemplation: “we have been talking about creating a protocol for responding to a dating violence incident” (20%); and preparation: “we have plans to implement a protocol for dating violence incidents in the next school year” (5%)]. Less than one in four (24%) school principals reported that they had a school protocol for responding to a TDV incident for the past year or more than 1 year (action and maintenance stages). School principals who had a school protocol reported assisting significantly higher numbers of victims of TDV in the past 2 years (M = 4.86, SE = ± 0.69) compared with those who did not (M = 2.67, SE = ± 0.27) [t = 3.50, df = 390, p = 0.001]. Similarly, school principals who had a school protocol reported a significantly higher number of perpetrators of TDV who received disciplinary action from school (M = 2.07, SE = ± 0.36) compared with those who did not (M = 1.08, SE = ± 0.21) [t = 2.70, df = 390, p = 0.007]. In addition, principals who had a school TDV response protocol reported significantly fewer barriers (M = 1.08, SE = ± 0.16) to assisting victims of TDV compared with principals who did not have a school protocol (M = 2.07, SE = ± 0.12) [t = 3.06, df = 389, p = 0.004].

School principals were asked to report if their school had certain practices and policies for prevention of TDV (Table 3). The majority of the principals reported that: their school did not provide formal training to staff on TDV issues in the past 2 years (62%), but their schools educated students about dating violence prevention (71%), healthy dating relationships (81%), and about where to report an incident of dating violence (77%).

Table 3.

School Policies and Practices for Preventing Teen Dating Violence

Item	Yes N (%)
School educates students about healthy dating relationships.	321 (81)
School educates students about where to report an incident of TDV.	305 (77)
School educates students about dating violence prevention.	281 (71)
School keeps TDV complaints in confidential file separate from academic records.	277 (70)
School provided training for staff on how to assist victims of TDV in the past 2 years.	150 (38)
School has a committee that meets periodically to address health and safety issues, which include TDV.	147 (37)
School's violence prevention/Safe School policy specifically addresses TDV.	138 (35)
School conducts periodic student surveys that include questions on TDV.	115 (29)
School has information posted about TDV that is easy for students to find.
Locations identified: Library, health class, cafeteria, hallways, hallway-outside social work office, social worker's office, counseling office, halls, on posters in guidance and in the halls, hallways and restrooms, family and consumer science classes, links to counseling services on website, all classes and guidance office, nurse's office, classrooms and SAP office, counseling office and restrooms, counselor's board and health classes, pamphlets near counseling office, life skills classes, student services office, student handbook, and bulletin board, we have info posted on: social media, harassment, violence etc.	119 (30)

Assistance to TDV victims

Principals were asked to report the number of TDV victims they assisted in the past 2 years. The average number of victims assisted was 3.28 (SE = ± 0.27) with a range of 0–30. The majority of the principals (57.0%) had assisted TDV victims in the past 2 years. In addition, principals were asked to report the number of TDV perpetrators who received disciplinary action from school within the past 2 years and the average was 1.04 (SE = ± 0.14) with a range of 0–15. More than a quarter of the principals (27%) reported sanctioning a disciplinary action for TDV perpetrators in the past 2 years. Principals who had formal training on TDV issues reported a higher number of victims assisted (M = 4.50, SE = ± 0.63) compared with those who did not (M = 2.63, SE = ±0.27) [t = 3.10, df = 390, p = 0.002]. Similarly, principals who had formal training on TDV issues reported a higher number of perpetrators receiving disciplinary action (M = 1.81, SE = ± 0.32) compared with principals who did not have training (M = 0.72, ± SE = 0.13) [t = 2.99, df = 380, p = 0.003].

Principals were asked to identify ways in which school personnel can help victims of TDV. The two most common choices were: referral of the victim to a school counselor (93%), and informing the victim's parents/guardians (85%). In addition, principals were asked to identify the barriers to assisting TDV victims by selecting options from multiple choices (Table 4). The two most common responses were: school personnel lack training to help TDV victims (43%) and shortage of staff (26%). A barrier score was computed for the respondents after combining all the barriers. For each barrier reported by the principals, a score of 1 was given (potential range = 0–7). The average barrier score for the principals was 1.57 (SE = ± 0.08), and the median number of barriers perceived by the principals was “one.” Principals who reported having a school TDV response protocol had a significantly lower barrier score (M = 1.08, SE = ± 0.18) compared with principals who did not (M = 2.70, SE = ± 0.09) [t = −2.81, df = 376, p = 0.03]. Principals who reported having formal training on TDV issues had significantly lower barrier scores (M = 1.03, SE = ± 0.13) compared with those who did not have training (M = 2.07, SE = ± 0.09) [t = −2.01, df = 372, p = 0.04].

Table 4.

Perceived Ways to Assist Victims and Barriers to Assisting Victims of Teen Dating Violence

Perceived ways
Referral of the victim to a school counselor	368 (93%)
Informing the parents/guardians of the victim	337 (85%)
Referral of the victim to legal authorities	293 (74%)
Reporting the incident to police	293 (74%)
Referral of the victim to a school nurse	238 (60%)
Referral of the victim to school social worker	230 (58%)
Reporting the incident to child protection agencies	226 (57%)
Referral of the victim to a physician/medical clinic	147 (37%)
Reporting the incident to teachers	135 (34%)
Help the victim obtain protective orders	107 (27%)
Other (please identify):	12 (3%)
Counseling services beyond school; referral to sexual assault advocate; referral to community support services; student assistance specialist; talk to teacher in faculty meetings; this is a parent/police issue, not a school issue.

Perceived barriers
Personnel in our school do not have the training to help victims of teen dating violence.	170 (43%)
Dating violence is a minor issue compared with other student health issues we deal with.	147 (37%)
Our school does not have enough personnel to help victims of teen dating violence.	103 (26%)
Parents would not approve of our school's involvement in helping victims of dating violence.	55 (14%)
It is not the school's responsibility to help victims of teen dating violence.	51 (13%)
Personnel in our school do not have the time to help victims of teen dating violence.	44 (11%)
The school administration does not want to deal with this sensitive issue.	16 (4%)
Other (please identify): We're here for academics, not dating problems; the trouble in finding out it is happening; it is not our business; lack of student reporting; limited resources; our judicial system is not proactive in addressing the needs of our students; our school lacks awareness about the issue; the school lacks ability to discipline, can only counsel and refer; staffing is short, but we make time; students don't identify themselves as victim/perpetrator to seek help; students might not feel the school can help; we are rarely informed of incidents; we help victims but do not have school sanctions turned over to police or School Resource Officer; would be great to have more personnel to help at this, but budget is the issue.	36 (9%)

N = 396.

A logistic regression analysis was conducted to explore the factors associated with principals' assistance to victims of TDV in the past 2 years (considered a binary outcome variable with assistance provided coded as “1” and did not assist victims coded as “0”). Six independent variables were found to be associated with whether a principal had assisted a victim in the past 2 years. A final model was created by adjusting these predictors for duration of employment, location of practice, race, gender, and age of the principals to calculate AORs for assisting TDV victims (Table 5). The highest odds of assisting TDV victims in the past 2 years were reported by principals in schools that had an active protocol to respond to TDV incidents (AOR = 2.64, 95% CI = 1.49–4.75, p < 0.001), and principals in schools that provided TDV training for staff and faculty within the past 2 years (AOR = 2.43, 95% CI = 1.45–4.00, p < 0.001).

Table 5.

Variables Predicting Whether Principals Assisted Teen Dating Violence Victims

Predictors	AOR (95% CI)
School has a protocol to respond to TDV incidents
No	Ref.
Yes	2.64 (1.49–4.75)^**
School provided training for staff on how to assist victims of TDV provided in the past 2 years.
No	Ref.
Yes	2.43 (1.45–4.00)^**
Responding principal received formal training to address TDV issues
No	Ref.
Yes	2.07 (1.24–3.54)^*
School keeps TDV complaints in confidential file separate from academic records.
No	Ref.
Yes	2.04 (1.21–3.34)^*
School's violence prevention/Safe School policy addresses TDV.
No	Ref.
Yes	1.96 (1.18–3.26)^*
School has a committee that meets periodically to address health and safety issues, which include TDV.
No	Ref.
Yes	1.72 (1.08–2.70)^*

= p < 0.01 and ^** = p < 0.001.

AOR, adjusted odds ratios for assisting TDV victims in the past 2 years (adjustments were made for duration of employment, location of practice, race, gender, and age of the school principals). CI, confidence interval.

Perceived roles of various groups in assisting victims of TDV

Principals were asked to rate their agreement with a series of statements on what should be the roles of school personnel in relation to preventing and responding to TDV. The majority of principals (93%) agreed that school counselors should play a major role in preventing and responding to TDV (Table 6). However, more than half (53%) of the respondents did not agree that school health teachers should play a major role in assisting the victims of TDV. In addition, about one in six principals did not agree that school personnel should work closely with school administrators to help formulate appropriate dating violence policies for students (15%) or that school personnel should be formally trained to assist students who are abused in dating relationships (13%) (Table 6).

Table 6.

Principals' Perceptions of Roles of Various Individuals in Assisting Victims of Teen Dating Violence

Groups	Major role N (%)	Minor or no role N (%)
School counselors	368 (93)	28 (7)
School administrators	282 (71)	110 (27)
School social workers	277 (70)	111 (28)
School resource officers (police)	273 (69)	123 (31)
School nurses	242 (61)	150 (38)
Health teachers	182 (46)	210 (53)
Students (peers)	143 (36)	244 (62)

Principals' beliefs about school personnel's role in preventing TDV
Perceived roles	Agree/strongly agree N (%)	Disagree/strongly disagree N (%)
One of the roles of school personnel is to cultivate the trust of students so that students report any occurrence of abuse in their dating relationships to school personnel.	384 (97)	12 (3)
Students who are abused in a dating relationship need to be encouraged to report the abuse to school personnel.	380 (96)	16 (4)
It is the responsibility of parents to work in collaboration with school personnel to prevent teen dating violence.	372 (94)	20 (5)
School personnel should be involved in developing the protocols that focus on how to respond to teen dating violence incidents.	348 (88)	36 (9)
School personnel should assist the victims of dating violence by referring them to legal authorities.	341 (86)	48 (12)
School personnel should be formally trained to assist students who are abused in a dating relationship.	337 (85)	51 (13)
It is the role of school personnel to work closely with school administrators to help formulate appropriate dating violence policies for students.	333 (84)	59 (15)

N = 396.

Discussion

This study is the first baseline assessment of current practices and perspectives of school administrators in relation to preventing and responding to TDV incidents. Several disconcerting findings warrant attention from schools and policy makers. The vast majority of schools did not have a violence prevention policy that specifically addressed TDV, and a majority of the school personnel were not provided training on TDV prevention within the past 2 years. In spite of this, the majority of the school principals reported interacting with victims and perpetrators of TDV. Without staff training, it remains unclear as to how school personnel can respond to TDV incidents and consider implementation of TDV prevention programs. In addition, absence of a specific protocol to deal with TDV incidents in a majority of the schools could mean that school personnel likely struggle to mount adequate and appropriate responses to TDV incidents. Another unique finding was that although the majority of principals had assisted a TDV victim in the past 2 years, the majority of the principals did not report having a provision of a disciplinary action for a TDV perpetrator in place. This indicates a potential gap in practice where all focus is on the victim, whereas perpetrators may not be actively deterred from future abusive behaviors.

When asked to identify the role that various groups of individuals should play in assisting TDV victims, the majority of school principals did not believe that health teachers and peers should play a major role. Further, the majority of the principals did not know that TDV victims frequently talk about the abuse experience with peers. In various studies of adolescent help-seeking behaviors regarding TDV, peers and teachers have been identified as the major and most trusted sources of help (Ashley and Foshee 2005; Martin et al. 2012). This is most concerning, as the role of teachers is central to implementation of any TDV prevention curricula. School administrators and policy makers need to consider the vital role that teachers can play in planning and implementing prevention programs involving students who may become victims or bystanders (Hickman et al. 2004; Johnson 2009; Leen et al. 2013; Martin et al. 2012; Offenhauer and Buchalter 2011).

Effective school violence prevention interventions emphasize the role of bystanders (Salmivalli et al. 2011; Twemlow and Sacco 2013). However, based on our study findings, it appears that the principals have become bystanders; and perpetrators of TDV will only do what the bystanders allow. This is further evident from the lack of principals' knowledge on TDV and the lack of specific school policies on TDV prevention. In addition, more than a quarter of principals in our study did not believe that school administration has a major role in assisting victims of TDV. Principals are prominent leaders in a school setting and should aggressively promote initiatives targeted toward TDV prevention and discourage bystander behavior at all levels in a school.

The results of this study suffer from all traditional limitations of observational and cross-sectional studies (e.g., socially desirable responses, recall bias, inability to illustrate cause and effect relationships, and self-reported data, just to name a few). The instrument was mostly a closed format and monothematic survey, which may be a threat to validity as we may have missed important perspectives of school principals. Finally, the response rate in this study was comparable to or better than the responses of school personnel in other studies (Dake et al. 2004; Price et al. 2016; Rickard et al. 2010). However, only to the extent that the non-respondents might have answered the questions differently could limit the external validity of the findings.

This study indicates a need for improvement in school-level practices on TDV prevention. Based on a comprehensive review of literature and practice guidelines from professional organizations, the following recommendations should be considered by schools for TDV prevention (CDC 2016; Freeman et al. 2013; Hickman et al. 2004; Johnson 2009; Khubchandani et al. 2013; Leen et al.2013; NASN 2013; Sherer and Nickerson 2010):

• First, periodic school-level surveillance of victimization and perpetration of TDV should be conducted to guide policies and TDV prevention practices.

• Second, periodic training of school personnel on TDV issues can help reach out to a greater number of victims and help reduce the prevalence of TDV.

• Third, development, documentation, and implementation of robust school violence prevention policies and periodically educating students and staff about these policies can help prevent TDV.

• Fourth, recent reviews and meta-analyses have found several school-based interventions to be effective in preventing TDV. Schools should consider implementing such evidence-based prevention interventions.

• Fifth, providing counseling, medical care, and referral to appropriate agencies and legal authorities can help victims of TDV and prevent repeated abuse.

• Sixth, majority of the states do not have a TDV-specific law that recommends schools to implement prevention practices. Individual states should consider enacting TDV prevention laws.

• Seventh, TDV is a complex problem with social, academic, and health-related impacts on adolescents; greater collaboration between schools, parents and pediatricians, community health and social services will be required for effective TDV prevention.

• Finally, organizations with concerns for school health and adolescents (e.g., National Association of School Nurses and American Academy of Pediatrics) can help bring greater attention to TDV and guide national initiatives to improve prevention practices and policies.

Conclusions

TDV affects a significant number of youth, and schools are now being looked at as an avenue for prevention and reduction of dating violence and similar violent youth risk behaviors. In this national study of high school principals, a state of affairs is presented on current practices and policies for TDV prevention. School principals identified a variety of barriers to assisting TDV victims, including lack of training and resources. Also, a significant proportion of schools do not have a protocol to respond to TDV incidents. This situation in U.S. high schools calls for greater emphasis on TDV prevention.

Footnotes

Acknowledgments

This study was funded in part by the Sponsored Projects Administration at Ball State University. Mrs. Cathy Whaley and the Northeast Indiana Area Health Education Center are funded in part by a grant award from the Health Resources and Services Administration (HRSA), U.S. Department of Health and Human Services.

Author Disclosure Statement

No competing financial interests exist.

References

Ashley

, Foshee

. (2005). Adolescent help seeking for dating violence: Prevalence, sociodemographic correlates, and sources of help. J Adolesc Health. 36, 25–31.

Centers for Disease Control and Prevention. (2016). Intimate partner violence: Prevention strategies. Retrieved from www.cdc.gov/violenceprevention/intimatepartnerviolence/prevention.html (accessed July 2017 ).

Dake

, Price

, Telljohann

, et al. (2004). Principals' perceptions and practices of school bullying prevention activities. Health Educ Behav. 31, 372–387.

DeLa-Rue

, Polanin

, Espelage

, et al. (2017). A meta-analysis of school-based interventions aimed to prevent or reduce violence in teen dating relationships. Rev Educ Res. 87, 7–34.

Fellmeth

, Heffernan

, Nurse

, et al. (2013). Educational and skills-based interventions for preventing relationship and dating violence in adolescents and young adults. Cochrane Database Syst Rev. 6, CD004534.

Freeman

, Rosenbluth

, Cotton

. (2013). Teen dating abuse: Recognition and interventions. NASN Sch Nurse. 28, 79–82.

Glass

, Fredland

, Campbell

, et al. (2003). Adolescent dating violence: Prevalence, risk factors, health outcomes, and implications for clinical practice. J Obstet Gynecol Neonatal Nurs. 32, 227–238.

Haynie

, Farhat

, Brooks-Russell

, et al. (2013). Dating violence perpetration & victimization among US adolescents: Prevalence, patterns, & associations with health complaints & substance use. J Adolesc Health. 53, 194–201.

Hickman

, Jaycox

, Aronoff

. (2004). Dating violence among adolescents: Prevalence, gender distribution, and prevention program effectiveness. Trauma Violence Abuse. 5, 123–142.

10.

Hoefer

, Black

, Ricard

. (2015). The impact of state policy on teen dating violence prevalence. J Adolesc. 44, 88–96.

11.

Howard

, Debnam

, Wang

. (2013). Ten‐year trends in physical dating violence victimization among US adolescent females. J Sch Health. 83, 389–399.

12.

Howard

, Wang

, Fang

. (2007). Psychosocial factors associated with reports of physical dating violence among US adolescent females. Adolescence. 42, 311–324.

13.

Howard

, Wang

, Yan

. (2008). Psychosocial factors associated with reports of physical dating violence victimization among US adolescent males. Adolescence. 43, 449–460.

14.

Johnson

. (2009). Improving the school environment to reduce school violence: A review of the literature. J Sch Health. 79, 451–465.

15.

Khubchandani

, Price

, Thompson

, et al. (2012). Adolescent dating violence: A national assessment of school counselors' perceptions and practices. Pediatrics. 130, 202–210.

16.

Khubchandani

, Telljohann

, Price

, et al. (2013). Providing assistance to the victims of adolescent dating violence: A national assessment of school nurses' practices. J Sch Health. 83, 127–136.

17.

Leen

, Sorbring

, Mawer

. (2013). Prevalence, dynamic risk factors and the efficacy of primary interventions for adolescent dating violence: An international review. Aggress Violent Behav. 18, 159–174.

18.

Martin

, Houston

, Mmari

, et al. (2012). Urban teens and young adults describe drama, disrespect, dating violence and help-seeking preferences. Matern Child Health J. 16, 957–966.

19.

National Association of School Nurses. (2013). School violence, role of the school nurse in prevention. Retrieved from www.nasn.org/PolicyAdvocacy/PositionPapersandReports/NASNPositionStatementsFullView (accessed June 2017 ).

20.

National Conference of State Legislatures. (2017). Teen dating violence. Retrieved from hwww.ncsl.org/research/health/teen-dating-violence.aspx (accessed June 2017 ).

21.

Offenhauer

, Buchalter

. (2011). Teen dating violence: A literature review and annotated bibliography. Washington, DC: US Department of Justice. Retrieved from www.ncjrs.gov/pdffiles1/nij/grants/235368.pdf (accessed June 2017 ).

22.

Price

, Khubchandani

, Payton

, et al. (2016). Reducing the risks of firearm violence in high schools: Principals' perceptions and practices. J Community Health. 41, 234–243.

23.

Prochaska

, Prochaska

, Levesque

. (2001). A Transtheoretical approach to changing organizations. Adm Policy Ment Health. 28, 247–261.

24.

Rickard

, Hendershot

, Khubchandani

, et al. (2010). School nurses' perceptions and practices of assisting students in obtaining public health insurance. J Sch Health. 80, 312–320.

25.

Rothman

, Xuan

. (2014). Trends in physical dating violence victimization among US high school students, 1999–2011. J Sch Violence. 13, 277–290.

26.

Salmivalli

, Voeten

, Poskiparta

. (2011). Bystanders matter: Associations between reinforcing, defending, and the frequency of bullying behavior in classrooms. J Clin Child Adolesc Psychol. 40, 668–676.

27.

Sherer

, Nickerson

. (2010). Anti‐bullying practices in American schools: Perspectives of school psychologists. Psychol Sch. 47, 217–229.

28.

Twemlow

, Sacco

. (2013). How & why does bystanding have such a startling impact on the architecture of school bullying and violence?. Int J Appl Psychoanal Stud. 10, 289–306.

29.

US Department of Education.. List of schools. Retrieved from https://nces.ed.gov/ccd/schoolsearch (accessed June 2017 ).

30.

Vagi

, Olsen

EOM

, Basile

, et al. (2015). Teen dating violence (physical and sexual) among US high school students: Findings from the 2013 National Youth Risk Behavior Survey. JAMA Pediatr. 169, 474–482.

31.

Vagi

, Rothman

, Latzman

, et al. (2013). Beyond correlates: A review of risk and protective factors for adolescent dating violence perpetration. J Youth Adolesc. 42, 633–649.