Introduction Domestic Violence has typically

Domestic Violence has typically been an issue assumed to be most relevant to

adults. However, the percentage of youth involved in violent dating relationships is

startling. Nationwide, nine percent of students have been hit or physically hurt on purpose

by their boyfriend or girlfriend (Center for Disease Control, 2012). Eight percent of

students have ever been physically forced to have sexual intercourse when they did not

want to (CDC, 2012). Overall, the prevalence of having been forced to have sexual

intercourse was higher among female (11.8%) than male (4.5%) students. Further, 20% of

adolescents report having experienced psychological abuse. Between 11% and 41% of

adolescents report using physical violence against their dating partner (Foshee &

McNaughton-Reyes, 2009).

Dating violence rates among youth vary from state to state. However, Nebraska

teens experience more dating violence than the national average. While 9 % (9.3% for

females and 9.5%for males) of adolescents nationwide had been hit, slapped or physically

hurt on purpose by their boyfriend of girlfriend (9th-12Th grade), 11% of Nebraskan

adolescents had been hit, slapped or physically hurt on purpose by their boyfriend or

girlfriend (11.1 for females and 10.7 for males). Nationwide, 8.0% (11.8 for females and

4.5% for males) of students had ever been physically forced to have sexual intercourse

when they did not want to while 10% of Nebraskan teens had ever been physically forced

to have sexual intercourse when they did not want to (11.1 % of Nebraskan females and

5.3% of males) (CDC, 2012). There are inconsistent statistics concerning dating violence

in adolescence due to the various definitions and measures used by researchers to assess

violence. Nevertheless, 11% is a substantial number of adolescents experiencing dating

violence. Dating violence in this investigation is defined as a pattern of physical, sexual

and/or emotionally abusive behaviors used to exert power and control over a dating

partner (Reed, Raj, Miller, Silverman, 2010). Further, the terms “adolescent dating

violence” and “teen dating violence” are used interchangeably. Teens who are involved

in a violent dating relationship in adolescence are more likely to be involved in a

subsequent dating violence relationship. Further, experiencing dating violence in

adolescence can put female and male teens at an increased risk for low self-esteem,

substance abuse, sexual risk behavior, suicide and they are more likely to enter into

another abusive relationship (Lewis & Fremouw, 2000). Additionally, victimized girls

are more likely to use unhealthy weight control methods (Lewis & Fremouw, 2000). A

five- year longitudinal study (Exner-Cortens, Eckenrode & Rothman, 2013), which

compared adolescents who experienced dating violence to a comparison group of

adolescents who had not experienced dating violence, found that female victims of teen

dating violence experienced substance abuse, depressive symptoms and interpersonal

violence as adults in greater percentages. The researchers found that males who

experienced teen dating violence experienced increased antisocial behavior, suicidal

ideation, substance abuse and adult interpersonal violence.

The negative impact of experiencing dating violence in adolescence endures into

adulthood. Adolescents who experience dating violence are more likely to experience

partner violence in adulthood. Relational patterns set in adolescence are likely to

influence adult relational patterns. Adult interpersonal violence has long been a

significant public health problem. One in four women will experience domestic violence

in their lifetimes. Eighty-five percent of all domestic violence victims are women. The

effects of experiencing domestic violence are vast and include depression, eating

disorders, anxiety, post-traumatic stress disorder (PTSD) and suicide. The costs of

domestic violence are not limited to the individual who experiences the violence, but to

the nation as a whole. In the United States the cost of intimate partner violence exceeds

$5.8 billion each year, $4.1 billion of which is for direct medical and mental health

services. Victims of domestic violence lost 8 million days of paid work each year,

because of the violence perpetrated against them by current or former husbands,

boyfriends and dates. “This loss is the equivalent of more than 32,000 full-time jobs and

almost 5.6 million days of household productivity as a result of violence (National

Coalition Against Domestic Violence, 2013, p.1).”

The negative effects of experiencing interpersonal violence are vast. Given that

the effects can be lifelong, prevention efforts to ameliorate the number of adolescents

experiencing dating violence are crucial. Prevention efforts can reduce the incidences of

dating violence and therefore reduce the physical, psychological and financial damage of

interpersonal violence (Foshee & McNaughton Reyes, 2009; Exner-Cortens, Eckenrode

& Rothman, 2013).

Adolescent/Teen Dating violence prevention programs have varied in length or

dosage. Prevention programs have ranged from one session to twenty sessions. There is

little empirical research about the most effective number of sessions for prevention efforts

to have the greatest impact, although there is evidence that a one session format does not

have a lasting impact on dating violence (Foshee & McNaughton-Reyes, 2009). Even

given the general consensus of the lack of effectiveness of one-session presentations, 35%

of respondents in a survey study by Weisz and Black (2009) carried out their prevention

programs in this format. The reason given by these investigators for only conducting one

session was the constraints imposed on them by the schools. Schools are pressured to

include in as much academic work as possible and to minimize any time considered to be

non-academic. Researchers often find it difficult to persuade school administrators and

teachers to allow them more than one session to provide programming to students. (Weisz

& Black, 2009).

Although a one- session format does not seem to provide lasting effects, there

may be some evidence that three sessions of treatment may be associated with positive

effects. For example, Jaycox and colleagues (2006) conducted a three-session curriculum

and showed changes in altering norms and behaviors regarding dating violence. Lavoie,

Vezina, Piche and Boivin (1995) evaluated the effects of short-term (i.e., 2 sessions)

versus longer-term (i.e., 4 sessions) dating violence prevention programs. The results

showed no differences between the long and short-term programs. Both groups showed

improvement from pre -to post-test in their attitudes and knowledge about dating

violence.

Along with time constraints, there are many financial constraints on schools and

prevention researchers. Long-term prevention programs cost more money than short-term

programs. Many prevention programs are provided by local domestic violence crisis

centers. These non-profit agencies have very little resources to spend on prevention.

There is a strong need for short-term effective prevention programs. An economic

evaluation of a dating violence prevention program’s cost-effectiveness is measured per

injury prevented or cost per life saved and subsequent cost of dating violence. A few

researchers have conducted an economic evaluation of dating violence programs. Life

Skills Training (Botvin & Griffin, 2004), a program aimed at preventing youth violence

by teaching social and emotional coping skills, saves $42.13 on future costs of violence

for every dollar spent on the prevention program. The Nebraska developed curriculum,

Reaching and Teaching Teens, costs twenty dollars. This program has yet to be evaluated

for cost-effectiveness (SAMSHSA, 2013).

Nebraska schools, as well as many nationwide schools, are not able to provide

much time for prevention programs due to the constraints placed on them to limit

nonacademic time. However, Nebraska is one of the several states that have adopted the

Lindsay Ann Burke Act. This law requires Nebraska schools to provide education to

teachers and students about dating violence (Nebraska Department of Education, 2010).

The Nebraska based curriculum, Reaching and Teaching Teens curriculum consists of 5

units. One of these units specifically addresses stereotypical gender norms. This unit’s

suggested length is from one to three sessions. A three-session format provides a thorough

program that focuses on defining, exploring gender norms and preventing violence

concerning gender norms (NDVSAC, 2011). The three sessions are formatted for a 40-

minute lesson that can be implemented independently or in conjunction with other lessons

(NDVSAC, 2011). Reaching and Teaching Teens has been used widely throughout the

state of Nebraska (Zinke personal communication, July 2013) but with little formal

evaluation. If shown to be effective, a three session dating violence prevention program

would meet the needs of the schools as required by the Lindsay Ann Burke Act. However,

determining what content is critical in such a short term prevention approach to dating

violence is very important given limited time and resources. Considering past research on

the content of existing prevention programs as well as the developmental needs of

adolescents related to dating are important considerations related to this issue.

Adolescence is a developmental time during which peers become extremely

important and interests in dating relationships emerge. Romantic relationships influence

“sexuality, intimacy and identity development; provide social support, influence secure

attachment; facilitate partner sorting and selecting and influence transformations in

family and peer relationships (Foshee & McNaughton-Reyes, 2009, p.141).” Further,

during adolescence, gender differences become more pronounced and adolescents rigidly

follow gender norms (Alfieri, Ruble, & Higgins, 1996). “Gender norms” are defined as

social expectations of appropriate roles and behaviors for males and females (Barker,

Ricardo, Nascimento, Olukoyo & Santos, 2010). Gender norms that promote male

dominance and control lead to greater violence against females and deleterious health

effects for males (Reed, Raj, Miller & Silverman, 2010). A strict adherence to female

gender roles is associated with a greater likelihood to be in an abusive relationship and a

greater length of time to leave an abusive relationship (Molidar & Tollman, 1998).

Violence prevention programs that have a focus on gender norms have been successful

(Foshee, 2005 &Barker et al, 2010) The combination of a strict adherence to gender

norms, great likelihood of entering into a dating relationship and high rate of adolescent

dating violence make adolescence an ideal time to provide education and prevention.

Flexibility in gender norms in adolescence can provide an adolescent with better

communication skills, which is essential for healthy relationships (Sullivan et. al, 2010).

Prevention efforts also need to be appropriately timed to take place when the topic

is relevant to the population and before the problem sets in. Dating violence peaks at age

16 (Foshee & McNaughton-Reyes, 2009). It would be most beneficial to enlist prevention

programs before this age. Nebraskan children and teens from ages11-17 have a higher

rate of dating violence than the national average (11% Nebraska vs. 9% national). Ages

11-17 are a crucial time in adolescents’ lives where they are developing interests in dating

relationships. Dating violence prevention programs need to be implemented to minimize

the detrimental effects of abuse. Prevention efforts should be implemented with youth

ages 11-17 due to the high prevalence of abuse in this population of Nebraska youth.

Purpose

Short-term gender norm focused prevention programs may hold great promise for

Nebraskan youth. These prevention efforts could contribute to the amelioration of the

devastating effects of dating violence in adolescence and into adulthood (Foshee, 1996).

In this study this researcher attempted to determine if a three session gender norms

module which is part of the larger Reaching and Teaching Teens intervention was capable

of modifying gender role beliefs. In particular, the hypothesis was: (a) Participants in a 3-

session Toxic Scripts intervention would demonstrate greater average stereotypical

gender norms attitude change than participants in a delayed treatment control group, and

(b) Differences would be maintained over a one-month post-treatment period.

CHAPTER TWO

REVIEW OF THE LITERATURE

“Gender norms or social expectations of appropriate roles and behaviors for males

and females” (Barker, Ricardo, Nascimento, Olukoyo & Santos, 2010, p.539) have a

strong influence on adolescent risky behavior. Extensive research on adolescent risky

behaviors has shown the role that gender role norms play; in particular, attitudes

accepting of traditional gender role stereotypes. Conforming to traditional masculine

stereotypes, whether one is male or female, is associated with an general increase in risk

taking, including driving without a seat belt (Raithal, 2003; Graine, 2009), drinking

alcohol (Schulte, Ramo & Brown, 2009), substance abuse (Schulte, Ramo & Brown,

2009; Barker et. al., 2010) and being arrested (Pulwerwitz & Barker, 2008). Further, the

adherence to the traditional gender role or stereotype is associated with a higher risk of

HIV and other STIs (Pulerwitz, Michaelis Verma & Weiss, 2010). Prevention efforts for

adolescent risky behavior have a long history, which began with targeting individual risk

factors for these behaviors but progressed to taking a broader approach to prevention

incorporating the influences of the cultural and social contexts. Adolescent substance

abuse prevention programs have recognized the importance that social norms including

gender role norms play in attitudes and behaviors recommending that a change in beliefs

about social norms as a crucial element for prevention programs (Flay, 2000; Flay &

Allred, 2001).

Prevention efforts for adolescent dating violence are relatively new and behind the

trajectory of prevention efforts of other adolescent risk behaviors. The majority of the

prevention programs have targeted individual risk factors in order to prevent dating

violence (e.g. child abuse, antisocial behavior, delinquency, attitudes accepting of dating

violence) (Foshee & McNaughton Reyes, 2009). These prevention efforts were rarely

systematically evaluated but rather were evaluated through unstandardized assessments

and anecdotal feedback (Weisz & Black, 2009). Currently, only three dating violence

prevention programs have been rigorously evaluated; Safe Dates (Foshee, Bauman, et

al’s, 2006;Foshee et al., 1998, 2000, 2004), The Youth Relationship Project (Wolfe et al.,

2003) and Ending Violence (Jaycox, 2006). Results from these evaluations showed

"theoretically or empirically based mediators of program effects based on program goals

(Foshee & McNaughton Reyes, 2009 p. 153).” Consistent with prevention for other

adolescent risky behaviors a theoretically based mediator is the role that attitudes

accepting of the traditional or stereotypical gender roles play in dating violence. Males

who adhere to stereotypical male gender norms are more likely to use abusive behavior

against women (Eisler et al., 2000; Parrot & Zeichner, 2003; Rando, Rogers, &

BrittanPowell, 1998; Robinson & Schwartz, 2004; Schwartz, Merta, Waldo, & Bloom-

Langell, 1998). Gender norms that promote male dominance and control lead to greater

violence against females and deleterious health effects for males (Reed, Raj, Miller &

Silverman, 2010). A strict adherence to female gender roles is associated with a greater

likelihood to be in an abusive relationship and a greater length of time to leave an abusive

relationship

(Molidar & Tollman, 1998.)

The influence of stereotypical gender role norms is consistent with adult research

on interpersonal violence (Jewkes, 2002). The role of stereotypical gender role norms in

adult interpersonal violence research has been discussed thoroughly. Specific gender role

beliefs have found to rationalize abuse against women. Beliefs based on “male privilege

and entitlement and the instrumental use of aggression to obtain dominance and control

are directly related to intimate violence (Schwartz, Kelley & Kohli, 2012, p. 1963).”

Boys and girls learn these beliefs through the familial and societal messages

communicated to them. Boys and girls learn to act in a way that is consistent with their

gender. Norms, values and beliefs about what it means to be female or male are

communicated to boys and girls through a social structure that reinforces male

entitlement and superiority and female subordination (Pleck, Sonestein & Ku, 1993 &

Dobash &Dobash, 1979; Finn, 1986; Schwartz, Kelley & Kohli, 2012).

In adolescence, gender-role norms become increasingly important and rigid.

Adolescents are learning appropriate behaviors for developing healthy romantic

relationships. These early relationships can influence future ideas of appropriate

gendered behavior in romantic relationships. Experiencing dating violence can interfere

with developing a healthy view of dating relationships. Therefore, learning to challenge

traditional stereotypical gender-role norms can prevent adolescents from entering in or

staying in an abusive relationship.

The importance of education about stereotypical gender norms and a need for

change in attitudes accepting of stereotypical gender norms cannot be understated in

violence prevention. A few dating violence prevention programs have addressed gender

role norms in their curriculum; Safe Dates (Foshee et al., 2006;Foshee et al., 1998, 2000,

2004), The Youth Relationship Project (Wolfe et al., 2003), Ending Violence (Jaycox et

al., 2006) and Reaching and Teaching Teens (NDVSAC, 1996, 2011). The Safe Dates,

Ending Violence and Reaching and Teaching Teens prevention programs were

schoolbased and focused on primary prevention. The Youth Relationship Project is a

community-based indicated prevention program for youth who are more likely to

perpetrate dating violence.) These prevention programs will be discussed in the

subsequent sections. For each program, the following will be discussed; the theory and

mechanisms through which each program hypothesizes to create outcomes, the

participants, the methodology, and how the program addresses gender role norms.

Safe Dates

The Safe Dates program is the most widely disseminated prevention program for

dating violence. This program targets behavioral outcomes through theoretically based

mediating variables (Foshee et al., 2005). “The Safe Dates program seeks to reduce

dating violence among participants by changing attitudes that condone partner violence

and promote gender stereotyping, building conflict management skills and increasing

knowledge about community resources (Foshee et al., 1998, p. 45).” These outcomes are

achieved through a ten-session interactive curriculum. Each session is approximately

forty-five minutes. The sessions include information on dating violence, caring

relationships, how to help a friend in an abusive relationship, gender stereotypes and

recognizing and dealing with feelings. The curriculum also includes a poster contest and a

thematic play about dating abuse.

Safe Dates curriculum works through a change in norms and learning conflict

management skills. Norms are “instruments of social control and thus have a significant

effect on behavior and conformity (Foshee et al., 2005, p.247).” An individual’s

likelihood to engage in a behavior is dictated by the perceived consequences of the

behavior and what those who are important to them would think of them (Fishbein &

Ajzen, 1975; Mizruchi & Perrucci, 1962; Solomon & Harford, 1984).

Norms that are accepting of dating violence and of traditional gender role norms

are changed through an increase in awareness of the negative consequences of dating

violence and enlisting appropriate peer responses to dating violence. Safe Dates

outcomes are also achieved through learning skills. Violence is often used when faulty

negotiating techniques are used that leave a conflict unresolved. Through the Safe Dates

curriculum, youth learn adaptive conflict management skills to avoid using violence to

communicate (Lloyd, 1987). Safe Dates was implemented in North Carolina with 14

public schools. The participants were stratified by grade (8th and 9th grade) and matched

on school size. One member of each matched pair was randomly assigned to the program

group, which consisted of 955 students, or the control group, which consisted of 1,010

students. The sample consisted of, 77% European Americans and 19% African

Americans and 51% were females. Pre-tests were given before the program and again

one-month, one year, two years, three years and four years after the program was

completed. The measures assessed specific physical, psychological and sexually abusive

behaviors. The mediating variables; dating violence norms, gender role norms, conflict

management skills and awareness of services were assessed. Dating violence norms and

gender stereotyping were measured through questions that asked how strongly

participants agreed or disagreed with certain statements (e.g. “A girl who makes her

boyfriend jealous on purpose deserves to be hit”; “In a dating relationship, the boy should

be smarter than the girl”). Conflict management skills were assessed by asking how often

participants used these skills in the last six months (e.g. “I told the person how I felt”).

Belief in the need for help was assessed by how strongly participants agreed or disagreed

with statements (e.g. “Teens who are victims of dating violence need to get help from

others.”) and asking if participants were aware of services for dating violence

(yes or no).

Adolescents in the Safe Dates prevention curriculum group reported less

psychological, moderate physical and sexual dating violence perpetrations and less

moderate physical dating violence victimization at all four follow-up periods compared

with the control group who did not have the Safe Dates curriculum. Further, those

participants in the Safe Dates group reported less acceptance of prescribed dating

violence norms, less acceptance of traditional gender norms, and greater belief in need for

help (Foshee et. al., 2005).

Safe Dates is a well-established prevention program for adolescent dating

violence, but the implementation of the program is costly. One of the major obstacles to

actually implementing a prevention program is the lack of time in schools and the lack of

resources. The Safe Dates program costs two hundred and twenty five dollars for the

curriculum. This does not include the cost of training staff and implementing certain

components of the program such as a poster contest and theatrical play. The publisher of

Safe Dates, Hazeldon, offers a one-day training for staff that costs $2,200 plus travel

expenses (Hazelden, 2013 via web). Thus, for many schools, implementing the program

with trained staff can be cost-prohibitive.

The Youth Relationships Project

Another program that has shown promising results from evaluations is The Youth

Relationships Project (YRP). This program is the only program evaluated for

effectiveness for at-risk adolescents (Wolfe et. al, 2003). The at-risk adolescents

identified for participation in the prevention program were adolescents who had a history

of abuse and maltreatment by their caregivers. These adolescents were referred from

Canadian Child Protective Services as the study was conducted in Canada. The YRP is

based on the theory of intergenerational violence (Cornelius & Resseguie, 2006) in which

maltreated children learn this abusive behavior from the behavior of their caregivers.

Social learning takes place by children watching the style of relationships their abusive

parents have and these representations/models are taken with them and re-played in their

dating relationships. The Youth Relationship Project sought to educate adolescents about

relationship abuse and power dynamics, build skills and increase social action. The

curriculum consisted of eighteen two-hour sessions that were interactive in nature through

guest speakers, role-playing and field trips. The education and awareness component

focused on the power imbalance in unhealthy relationships and awareness of one’s own

power. This education is based on the Duluth model of adult domestic violence (Pence

&Shepard, 1999), where power and control are seen as the drivers of abuse. The skill-

building component aided youth in developing healthy communication skills such as

listening, empathy, and assertiveness. The social action component included education

about resources in the community, visiting a local social service agency and fundraising

for community needs. The group was co-facilitated by a female and male. This gender

dynamic was implemented in order to model healthy effective communication between

the two genders to the participants.

The program evaluation research consisted of 158 adolescents who were randomly

assigned to either the prevention group (n=96) or a control group (n=62). Data was

collected at baseline, four months after the prevention program began, after the program

ended and then again every other month for a total of seven data collection periods.

The majority of the participants were Caucasian (85%) while 8% were First

Nation, 4% Asian and 4% African Canadian. The program’s effectiveness was assessed

by contacting the adolescents twice a month. If a participant indicated that he or she was

involved in a dating relationship the participant would complete a questionnaire about

dating violence. All youth were assessed sixteen months after the program ended.

Participants in the prevention program showed significant reduction compared to the

control group over time in physical dating abuse perpetration, emotional dating abuse

victimization and victimization from threatening behaviors. The prevention group

compared to the control group did not show a reduction in emotional abuse perpetration,

threatening behaviors toward dates or physical dating abuse victimization. The program

was more effective for boys than girls in reducing victimization from physical abuse.

This study was more inclusive in their definition of dating violence, including

more than just physical abuse. They included emotional abuse and threatening behaviors

as different categories of abuse. Youth Relationships Project was designed to change

behavior through mediating variables. However, Wolfe et al. (2003) did not assess many

of these mediators in their analysis. The prevention program was hypothesized to

“increase knowledge about dating violence, alter attitudes towards dating abuse, increase

understanding of power dynamics in relationships, and improve various communication

and problem solving skills" (Wolfe et al., 2003, p. 282). Wolfe et al. (2003) only assessed

the effects of three of these hypothesized mediators. The prevention program was

associated with a reduction in trauma symptoms and had no effect on problem solving

skills and hostility. The process through which the intervention is expected to lead to

changes in outcomes is not fully known because a thorough analysis of all the mediators

was not completed.

This study showed some promising results but limited generalizability because the

sample selected were all youth who had been maltreated. These results may not be

transferable to youth who have not experienced abuse. The program was also more

effective for males than females in reducing victimization from physical abuse. There is

disagreement in the dating violence literature about whom experiences dating violence

more, males or females (Mulford & Giordano, 2008; Reed 2008; White, 2009 & Foshee

& McNaughton-Reyes, 2009). Some studies estimate that boys and girls experience

dating violence equally while other studies have girls perpetrating more abuse and some

have boys perpetrating more abuse. The difference in statistics is due to the measurement

used. When the questions are asked within a context of self-defense, more girls state they

use violence as a means to defend themselves from abuse. Although researchers cannot

agree on the validity of perpetration and victimization statistics, statistics show that girls

experience more serious injuries and deaths due to dating violence than boys (Reed 2008;

White, 2009). Therefore, programs need to be equally effective program for boys and

girls in order to reduce the number of dating violence injuries and deaths (Mulford &

Giordano, 2008; Reed 2008; White, 2009 & Foshee & McNaughton-Reyes, 2009).

The Youth Relationships Project’s eighteen-session format for a teen prevention

program is an extensive time commitment for the youth, their caregivers and the

prevention workers. A longer length of the sessions increases the cost of the prevention

program. Nebraska schools only require a few days for dating violence education. An

eighteen-session format would be costly and difficult to implement given the length and

time commitment. Results from the YRP program show that attending more sessions did

not increase the level of change in the participants. Participants who attended more

sessions did not differ from participants who did not attend all sessions. The process

through which the program is hypothesized to lead to suggested outcomes is through an

increase in understanding of power dynamics in relationships and an increased awareness

of gender stereotypes. This is consistent with research from the Safe Dates program,

where awareness of gender stereotyping was shown to be a mediator between the

program and the outcomes. Awareness of power dynamics in a relationship has shown to

change intimate partner violence in adults. Gaining awareness of the power and control

utilized in abusive relationships has shown to decrease adult intimate partner violence

both on perpetration and victimization (Pence & Shepard, 1999). Wolfe and associates

(2003) did not measure participants’ attitudes accepting of stereotypical gender norms;

therefore it is impossible to ascertain if education about gender norms produced any of

the changes they reported.

The YRP was carried out in the community rather than a school. This is a

promising approach because schools are difficult to access and it is difficult to commit

students for an extended period of time for non-academic activities. Community centers

and domestic violence crisis centers are great places to hold prevention programs because

there are fewer restrictions placed on the researchers. Domestic violence crisis centers

have the knowledgeable staff and resources that would be able to teach and implement

prevention programs effectively (Weisz & Black, 2009).

Ending Violence

The Ending Violence prevention program was developed by the organization,

Break the Cycle. Ending Violence focuses on preventing dating violence among Latino

teens. This curriculum was created because there is a lack of dating violence prevention

programs that focus on ethnic minorities. The program is based on Social Learning

Theory (Bandura, 1973 &1977) which theorizes that youth who witness or experience

domestic violence in their home will learn that violence is an acceptable way to solve

conflicts in their lives and transfer this to dating relationships. Further, media and cultural

factors can teach children that violence is an acceptable way to deal with conflict in a

relationship (Jaycox et. al., 2006).

The Ending Violence curriculum sought to “alter knowledge and norms about

dating abuse, increase favorable attitudes for seeking help for dating violence and

decrease dating abuse perpetration and victimization (Jaycox et al., 2006, p.694).” The

program was designed to achieve these goals through altering participants’ family, peer

and cultural norms by role modeling acceptable alternatives to violence for solving

conflict. This was conducted through teacher education and role modeling. The Ending

Violence program focused on Latino youths, many who were immigrants. Cultural norms

of the United States and the cultural norms from their culture of origin were discussed.

Teachers, who are all attorneys, modeled that violence is common but not acceptable.

This was done by providing information about individual rights, in particular, laws that

punish perpetrators of violence and legal resources and measures victims of dating

violence can use to protect and empower themselves. Further, youth act out a court

procedure that enables them to gain knowledge of laws in action as well as provide them

with assertive help-seeking behavior. Lastly, the program included a peer leadership

program where participants were educated and empowered to speak publicly in their high

schools and colleges about dating abuse. An aspect of the peer leadership program also

included a barter program where youth could volunteer with Break the Cycle agencies for

free legal services. (Jaycox et al., 2006).

Bilingual teachers and attorneys administered the program over three sessions.

Eleven schools in Los Angeles were randomly assigned to the prevention program group

or a standard health curriculum. There were 2540 students who participated with 80% of

the students identifying as Latino. A pre- and post-test were given as well as a follow-up

test six months after the program was completed. The results showed that the program

had significant effects in increasing student knowledge about laws related to dating

violence, decreasing accepting attitudes about female-on-male violence, and increasing

the likelihood of seeking help immediately following the program but faded after six

months. The only intervention effects that lasted after six months were knowledge about

dating violence laws and the usefulness of seeking help from an attorney for dating

violence. There was no difference between treatment and control groups in acceptance of

male-to-female violence, abusive dating experiences, dating violence perpetration and

victimization. An interesting finding in this study was that attitudes accepting of

femaleon-male violence changed (their attitudes became less accepting of female-on-male

violence) due to the program but attitudes accepting of male-on-female violence did not

change. The authors stated that the reason for this might be because male-on-female

violence was already viewed as strongly negative while female-on-male violence was not

viewed as negative (Jaycox et al., 2006).

The results of this evaluation indicate that Ending Violence had limited success. A

thorough analysis was not completed on many of the program components that were

hypothesized to lead to suggested outcomes. “Only student knowledge about attorneys,

proclivity to seek help and perceived helpfulness of these professionals showed lasting

changes in this program (Jaycox et al., 2006, p.703).” The program focused a majority of

the time on the legal rights of individuals involved in dating violence and resources they

could obtain. The program briefly discussed cultural norms that promote violence but did

not spend a great deal of time on this issue and did not specify stereotypical gender norms

as one of the factors of dating violence. Attitudes accepting of stereotypical gender norms

have been associated with dating violence (Eisler et al., 2000; Parrot & Zeichner, 2003;

Rando, Rogers, & Brittan-Powell, 1998; Robinson & Schwartz, 2004; Schwartz, Merta,

Waldo, & Bloom-Langell, 1998). A curriculum, which focuses solely on the

consequences of dating violence, does not address the underlying social norms.

Ending Violence curriculum did a good job in enlisting culturally competent

professionals for an advisory board. The advisory board consisted of mental health

professionals, doctors, social workers and teens, all of who had experience and

knowledge working with Latino youth and understood the needs and realities of the

culture. Jaycox, Arronoff and Shelley (2006) worked with the advisory board to make the

curriculum as culturally appropriate as possible for Latino youth. Many authors of

prevention programs are advised to reach out to the needs of the population to which they

want to serve, but few do. Enlisting the population that you want to study into the

creation of the curriculum allows for greater utilization rates and the likelihood of

resonating and changing attitudes (Creswell, 2013).

Reaching and Teaching Teens

Reaching and Teaching Teen curriculum was developed by the Nebraska

Domestic Violence and Sexual Assault Coalition (NDVSAC, 2011) in response to the

higher than national rate of dating violence in Nebraska youth. This program was

developed using a combination of the Ecological Systems Theory (Bronfenbrenner, 1979)

Feminist Theory (Walker, 1990) and Social Learning Theory (Bandura, 1973)

(Zinke personal communication, July, 2012).

Ecological Systems Theory (Bronfenbrenner, 1979), states that an individual does

not exist in isolation but rather is influenced by the systems in one’s environment. These

include familial, cultural, societal, economic and political systems. An individual is

influenced by one’s environment and influences one’s environment. The curriculum

addresses the many systems that influence and contribute to dating violence. These

include family history of domestic violence, cultural norms, societal norms laws and

policies protecting adolescent victims of dating violence (NDVSAC, 2011). Social

Learning Theory is demonstrated in the curriculum by addressing the influence that

society, family and peers have on an adolescent’s behavior. Social Learning Theory,

(Bandura, 1973) when used to describe violence, describes the process of learning violent

attitudes and behaviors. Children who witness violence learn that this is the acceptable

way to solve conflicts. Children learn the functional nature of violence as an acceptable

way to express themselves, to solve and to control and dominate another person. Images

of violent relationships in the media) are frequent in society and desensitize individuals to

violence, especially children who have been victims or have witnessed relationship

violent at home. These children learn the violent behaviors in their homes and then it is

reinforced by the violent images they see in the media. The violent sexual relationships

portrayed are very appealing to youth who know little about relationships and see this

violence as normative behavior in a relationship. The violent behavior becomes a means

to solve conflict and meet the adolescent’s need to be accepted. (Werkerle & Wolfe,

1999).

Reaching and Teaching Teens curriculum also incorporates Feminist Theory.

Feminist theory of violence views the social context within which individual acts of

violence occur. Violence against women occurs within patriarchal societies. Patriarchal

societies reinforce male entitlement and superiority. Violence is used to maintain power

and control over another person. Intimate partner violence is patterned along gendered

lines where females are the majority of the victims. Feminist efforts to combat violence

against women are aimed at awareness of the social environment that reinforces male

entitlement and superiority and women’s oppression. This awareness and knowledge is

transferred to empowerment of women. Empowerment is used to take back control of

one’s environment and to advocate for change (Hunnicutt, 2009).

The authors’ goals for the Reaching and Teaching Teens program are to increase

awareness about dating violence, increase knowledge about the effects of dating violence,

increase knowledge of the services for victims of dating violence, decrease attitudes

accepting of dating abuse, decrease dating abuse, change attitudes accepting of

stereotypical gender norms and empower youth to speak out against abuse (NDVSAC,

2011). The curriculum includes information about knowledge, attitudes and behaviors.

The authors included all three variables due to the complex relationship between

knowledge and attitudes with actual behavior (Weisz & Black, 2001). Knowledge,

attitudes and social norms all influence behavior. The rationale for changing attitudes is

that it sheds light onto likely future behavior (Kahle, 1984). In particular, changing

attitudes about stereotypical gender norms can influence dating violence behavior

(NDVSAC, 2011).

Reaching and Teaching Teens is designed for youth in grades 6-12. The

curriculum consists of 5 units; Sticks and Stones: Sexual Harassment; Every 8 Seconds:

Sexual Assault, When Flowers Don’t Fix It: Dating Relationships and Toxic Scripts:

Gender Violence. The curriculum combines lesson plans, activities, role-plays, teen

survivors’ experiences and suggested resources. The curriculum units can be presented

individually or in combination with any unit. Each unit contains goals and objectives and

sample lesson plans for various time amounts.

The unit that specifically addresses stereotypical gender norms is Unit 5: Toxic

Scripts. This unit’s suggested length is from one to three sessions. A three-session format

provides a thorough program that focuses on defining, exploring gender norms and

preventing violence concerning gender norms (NDVSAC, 2011). The three sessions are

formatted for a 40-minute lesson that can be implemented independently or in

conjunction with other lessons (NDVSAC, 2011). The goals of this unit are to increase

awareness of gender norms, knowledge of gender norms and decrease attitudes accepting

of stereotypical gender norms and increase gender flexibility.

Reaching and Teaching Teens has been used widely throughout the state of

Nebraska (Zinke personal communication, July 2013) but with little evaluation. To date

there is only one published evaluation of the curriculum. This was completed in a

predominantly African American (99%) middle school population in Detroit Michigan.

Weisz & Black (2001) chose the curriculum for its thorough attention to dating violence,

gender norms and the culture sensitivity of the program. The curriculum was conducted

in twelve, one and one half hour sessions over a six-week time period. Two hundred and

fifty students participated in the study; half to the intervention and the other half was used

as a wait-list control group. The program was conducted after school by facilitators from

the local sexual assault and domestic violence crisis center. Participants were given

questionnaires at the start of the intervention; right after the intervention was completed

and six-months after the intervention had ended. The questionnaires were used to measure

participants’ knowledge, attitudes and behaviors. Knowledge of sexual assault and dating

violence increased at the end of the program and was maintained at six-month follow-up

as well as an improvement in attitudes of sexual assault and dating violence (decreased

acceptance of dating violence and sexual assault). Actual behavioral data were not

collected. A limited amount of feedback was received from the participants about what

aspects of the program were most helpful. The authors of the evaluation did not conduct a

mediation analysis to see what aspects of the program may have led to a lasting change in

the participants (Weisz & Black, 2001).

The Reaching and Teaching Teens curriculum is unique in its theoretical basis.

Feminist theory is one of the central theories used in the creation of the curriculum.

Prevention programs for adolescent dating violence have largely ignored the gendered

nature of violence. Many adolescent dating violence prevention programs report that girls

and boys use violence at equal rates. This belief comes from the use of no contextual

measures where “respondents are asked to report on violence that ranges from slapping

each other to using a lethal weapon.” (Hunicutt, 2009, p. 557) The weaknesses of these

measures are that do not take into account the social environment within which these acts

occur. When the social context is taken into account, women often report using violence

in self-defense and men use more lethal methods of violence (Dobash & Dobash,

1992;Dobash & Dobash 2004).

A weakness of the Reaching and Teaching Teens curriculum is that it has not been

extensively evaluated even though it is used widely throughout the state of

Nebraska. Further research needs to be completed on the effectiveness of the curriculum.

Limits of Prevention Programs

The four programs aforementioned have shown some promising results for the

prevention of adolescent dating violence but further research needs to be completed.

Attitudes accepting of stereotypical gender role norms are associated with many

adolescent risky behaviors including dating violence. Gender norms should be viewed

through a Feminist lens. Feminist theories have explained violence against women for

many years (Hunnicutt, 2009). Violence occurs along gendered lines and ignoring this

reality is ignoring that “acts of violence against women are embedded in a larger social

organization (Hunicutt, 2009, p. 556).” This is underscored in an article by Barker,

Ricardo, Nascimento, Olukoya and Santos (2010) in which they reviewed 54 evaluation

studies of programs with men and boys for prevention and interventions in reducing

violence against women and HIV. Barker et al. (2010) found that programs that use a

gendered framework in their programs are more effective in producing behavior change.

Further evaluation studies of the programs in existence need to be conducted. Safe

Dates is the only program in which the authors conducted a mediation analysis to find the

process through which the program worked. One of the mediators was through a change

in attitudes accepting of stereotypical gender norms. The Youth Relationship Project

prevention program suggested that education about stereotypical gender norms would

lead to a change in dating violence behavior but the authors did not conduct a mediation

analysis to test their hypothesis. The Ending Violence program hypothesized that one way

through which their program would work was through a change in cultural and societal

norms accepting of violence. The authors did not conduct a mediation analysis of their

hypothesis and the education about cultural and societal norms accepting of violence did

not specifically include information about stereotypical gender norms (Foshee &

McNaughton-Reyes, 2009). The Reaching and Teaching Teens curriculum uses a

Feminist framework for prevention efforts including education about gender norms.

There has been only one program evaluation completed in which participants’ attitudes

about stereotypical gender roles were not assessed. Further research concerning the

effectiveness of addressing gender role norms in dating violence prevention needs to be

completed.

Prevention programs need to be short term in order to encourage schools and

community centers to use the program. Nebraska schools are required by the Lindsay

Ann Burke Act (Nebraska Department of Education 2010) to provide dating violence

education, but this education usually comes in the format of a one to three day

informative session by a staff of the local domestic violence shelter. A study by Weisz and

Black (2009) indicated that 35% of participants carried out their prevention programs in a

one-session format. The main reason for this chosen format was due to schools placing

constraints on time spent with students. The No Child Left Behind Act (2002) stressed the

importance of academic performance. Although, this is no longer the law of the land,

schools still implement and emphasize yearly testing of student knowledge. This attention

to testing is often to the detriment of attending to social and emotional needs.

When budgets are squeezed the first thing to go is counselors and interventionists.

Legislators fail to make the connection between emotional health and academic success

(Weir, 2012). Further, a longer length of a prevention program is not essential for change.

Short-term programs have showed promising results (Jaycox et. al., 2006; Lavoie, Vezina,

Piche & Boivin, 1995). Further research is needed to find the optimal amount of sessions

or the least number of sessions needed to produce change that cannot be furthered by

more sessions.

CHAPTER THREE

METHODS

Purpose of the Study

The purpose of this study is to add to the dating violence prevention program field

by implementing a dating violence prevention program that addresses stereotypical

gender norms, that is appropriate for Nebraska youth and is short term. Specifically, the

study addresses the following questions: (a) Do participants in a 3-session Toxic Scripts

intervention demonstrate greater average stereotypical gender norms attitude change than

participants in a delayed treatment control group, and (b) are differences maintained over

a one-month post-treatment period.

Participants

A group of adolescents in grades 9-12 were recruited for this study. This grade

range was selected because Nebraska has a higher than national rate of males and females

in 9th through 12th grade who report experiencing dating violence within the last twelve

months (CDC, 2012). Participants were adolescent at the Lighthouse, an after-school

program in Lincoln, Nebraska.

Of importance in this study is the effectiveness of the Toxic Script lesson with

adolescents in Nebraska. Nebraska is a conservative mostly rural state in the Midwest. Of

the nearly 1.7 million individuals in Nebraska, 90% are White, 5.5% are Hispanic, 2%

Asian 4% are African American, 1% are Native American and those reporting two or

more races is 2% (Nebraska Department of Health and Human Services, 2010). This

researcher attempted to enlist a similar demographic makeup for the sample in this study.

Among the entire sample, the participants were 34.8% Black/African/American, 26.1%

more than one ethnicity, 17.4% White/Caucasian, 8.7% Asian/Pacific Islander, 4.3%

Puerto Rican and 8.7% not reporting their ethnicity. The experimental group had 38.5%

Black/African-American, 23.1% more than one ethnicity, 15.4% White/Caucasian, 7.7%

Puerto Rican, 0% Asian/Pacific Islander and 15.4 did not report. The wait-list control

group had 30% Black/African American, 30% more than one ethnicity, 20%

White/Caucasian, 20% Asian/Pacific Islander, 0% Puerto Rican and 0% did not report.

The participants were originally planned to be matched in pairs based on age and

then randomly assigned to either the treatment or wait-list control group. Dating violence

increases with age and peaks at age 16 (Foshee, McNaughton-Reyes & Wyckoff, 2008).

Matching the participants based on age would ensure that there were equal numbers of

participants within each age range. This would control for the likelihood that this

variable would affect the outcome measure. This could not be done due to the low sample

size. This researcher had to place the participants in the group that fit best with their

schedule.

A child assent form was used in order to explain the study to the participant and

ensure that participants were willingly participating. This form stated that the

adolescent’s parents or legal guardian said that they could participate in the research

project but the researcher would like the adolescent to also make the decision. The child

could say yes or no. This helped control for the influence from the parent and/or legal

guardian that the child may feel to complete the program.

Eligibility

The guidelines that were implemented in this study included any youth in 9th-12th

grades.

Sample Size

In a meta-analysis of dating violence prevention programs among middle and high

school students, the overall effect size for changing attitudes was .325 (r=. 325) (Ting,

S.R, 2009). This researcher expected a greater change among the treatment versus

comparison participants therefore; this researcher used a directional, one-tail, t-test. This

researcher set the alpha level at .05 (p<0.05). To obtain an effect size of this magnitude,

this researcher needed 60 participants total, 30 in each group, to obtain medium effect

(Faul, Erdfelder, Lang, & Buchner, 2007). This was likely unrealistic because of limited

resources of this researcher. This researcher attempted to recruit as many participants as

possible.

Recruitment

Fliers including information about the dating violence prevention program were

posted around the Lighthouse. This researcher met with the Lighthouse staff many times

to enlist their support for the study.

Design of Study

It is important for a clinical trial to have a control group to achieve increased

internal validity so that causal inferences can be drawn about the effects of the program.

This researcher attempted to recruit enough participants for both groups and use a random

number table to assign participants to either treatment or wait-list control group after

matching participants by age. This could not be done due to the low sample size.

This researcher had to place the participants in what group fit best with their schedule.

This researcher had a wait-list comparison group. The wait-list group and the

treatment group were both assessed at the same time: pre-treatment, post-treatment and

follow-up at one month. Dating violence prevention treatment effects usually are present

at the end of the program, but sometimes fade at the follow- up assessment.

Wait-list control groups allowed this researcher to assure that the two groups were

similar in regard to developmental changes and exposure to specific events. Further, it

increased the likelihood of having changes in gender norms that are due to the Toxic

Scripts lesson, rather than due to common threats to internal validity like history,

maturation, testing, regression to the mean, and selection factors (e.g., age). The wait-list

control group is ethical in this study because withholding treatment would be to the

detriment of that group due to the high chances of participants in this age group who may

have experienced dating violence.

Treatment Protocol

The Reaching and Teaching Teens protocol was written and developed by the

Nebraska Domestic Violence and Sexual Assault Coalition. The Toxic Script lesson was

written to be anywhere from one to three sessions long. The program is somewhat

flexible because it was developed to be used in both community centers and in schools.

This researcher conducted the prevention lessons at the Lighthouse, which allowed for

flexibility to make the lesson three sessions in length. In a school setting it is harder due

to academic and regulation constraints in the local schools. Originally, an objective

observer was planned to be used to measure how well this researcher adhered to the

treatment protocol (treatment fidelity). This data would have allowed this researcher to

analyze the results to see if a difference in this researcher’s level of adherence to the

lesson affected the participant outcomes (e.g. lower scores on fidelity checklist could be

associated with smaller changes in gender norm attitudes).

The program was three sessions long. Effective short-term (three sessions) dating

violence prevention programs have been implemented by Jaycox and colleagues (2006).

Due to developmental changes adolescents’ attention span is relatively short (average

middle school students attention span is 10-12 minutes (Strauch, 2003). A short program

would more likely keep the participants’ attention than a longer prevention lesson.

Measures

This researcher had participants fill out a demographic questionnaire. This

questionnaire included questions regarding age, gender, ethnicity, school, grade level,

family composition and sports/afterschool activities.

Among the adolescent dating violence prevention literature there is little

consistency in what measures are used and what is actually being measured. Many of the

programs simply use a few homemade questions to assess attitudes and behaviors. Other

authors use instruments that only measure physical dating violence failing to account for

emotional and sexual violence. Based on the recommendation from the Center for

Disease Control Compendium of assessment tools for measuring violence-related

attitudes, behaviors, and influences among youth, this researcher selected to use the

instrument discussed below (Dahlberg, Toal, Swahn, & Behrens, 2005).

In order to measure the stereotypical gender norms the Attitudes Towards Women

Scale-Adolescent was administered (Galambos, Petersen, Richards, & Gitelson, 1985;

Dahlberg, L.L., Toal, S.B., Swahn, M., & Behrens, C.B., 2005). This instrument is used to

measure gender stereotyping in 11-18 year olds. It was normed on adolescents in grades

6-12 from various geographic locations and socioeconomic backgrounds in the United

States. Each question has four response options, ranging from “agree strongly" to

"disagree strongly." The Cronbach alpha of the 15-item form has been found to be 0.89.

According the Galambos (personal communication, December 1, 2012), caution should

be taken in using this measure because girls are more sensitive to questions on these

measures than boys.

This measure is suitable for use with ethnic minority participants because it has

been used before in studies with diverse populations. Thomas and Hersen (2011) suggest

that when choosing a measure for one’s study one should make sure it is standardized and

normed on the particular ethnic group in one’s study. Further, participants were selected

that are fluent in the English language to eliminate any wrong measurement due to the

assessment not being linguistically appropriate. Great care was taken in interpreting the

results because the definition of stereotypical gender norms may be Eurocentric. Different

cultures may view gender norms differently than European Americans. For example,

Hispanics are more likely to adhere to traditional gender roles. The adherence to these

traditional gender roles should be interpreted through the lens of the Hispanic culture.

Further, African American women tend to have less stereotypical gender role norms.

These differences should be interpreted through a cultural lens and not viewed as

deficient because it differs from the European American view on gender roles (Sue &

Sue, 2011).

Anticipated Ethical Issues

To minimize the general ethical concerns for research with this sensitive and

vulnerable population, this researcher followed general regulations determined by the

Institutional Review Board of Loyola University Chicago and the American

Psychological Association. The participants were especially vulnerable populations

because they are children. The legal age of consent in Nebraska is 19 years old. Anyone

younger than that age must have their parents give consent in order for them to

participate. What the child may want and what the parents may want may differ. Abiding

by the ethical principal of autonomy, participants signed a written assent form agreement

in order to participate (Thomas & Hersen, 2011). A child assent form is recommended for

children age seven and above. A child assent form allowed the child to determine if they

wanted to participate. If a child did not sign the assent form, this researcher did not

include the child as a participant.

Confidentiality was discussed and reinforced throughout the research process with

the participants and their parents. Participants were given alias names in order to protect

their information. Additionally, all study material was seen only by myself and was kept

locked in a private filing cabinet. Information concerning the limits of confidentiality was

addressed with participants and their guardians. Participants were aware that

confidentiality cannot be maintained if a child stated that they are a threat to themselves

or others and/or if a participant disclosed that an adult is abusing them. This did not

occur in the study. If these circumstances, did occur this researcher would report it to

Child Protective Services of Nebraska in abidance by the mandatory reporting law.

Further, this researcher met with the participants and their guardians to stress that the

prevention program will be most helpful if guardians allowed their children to participate

with minimal interruptions (e.g. asking about specific information a child provides in the

group). This researcher ensured that participants understood the limits of confidentiality

by discussing this during the first session.

Participants received informed consents that were discussed with the participants

on all pertinent information points as well as benefits and risks or discomforts one may

encounter with participation in this study. There is much concern about participants’

negative emotional reactions to participating in dating violence research. While this

concern is warranted, studies have shown that participants report positive benefits from

contributing to dating violence research including insight into one’s life and personal

relationships (Shorey, Cornelius Bell, 2011). The benefits of participating in dating

violence research may outweigh the risks (Shorey, Cornelius Bell, 2011; Newman,

Willard, Sinclair, and Kaloupek, 2001).

Although risks or discomforts appeared to be minimal, there were two

considerations this researcher strived to be aware of and attempted to minimize while

working with this population. These considerations were: 1) on occasion participants may

experience temporary distress when discussing gender stereotypes and violence 2) due to

the inevitable power differential between participant and researcher, the participant may

have felt pressured to answer all assessment questions and disclose information they are

uncomfortable sharing. Specific ways used to counter these risks were; if a participant felt

distressed at any time during the study, they had the right to skip the question or session,

were encouraged to let this researcher know in order to provide them with resource

information for their community’s mental health clinic, or they could withdraw from the

study at any time without any consequences. Ultimately, this researcher attempted to

utilize caution at all stages of the study to ensure the participants felt comfortable/safe

and non-coerced.

Planned Data Collection

As suggested by Thomas and Hersen (2011) one should plan the organization of

their data collection through a work plan. Adhering to this advice, this researcher had a

three ring binder that was separated into sections of the study. Some of these sections

included; a description of the study, a copy of the instruments, a codebook and a copy of

the raw data (Thomas & Hersen, 2011). The codebook contained information about

decisions to be made. For example, a protocol for when a participant skipped a question

on the assessment or circled two responses.

This researcher verified the data entry for accuracy. To manage for mistakes that

may be made, a sample of the participants’ data was checked. This included the first two

assessments, the middle two assessments and the last two assessments.

This researcher included in the data, participants who have completed the pre-test

and post-test measures as well as having attended every prevention program session. This

researcher tested whether attrition created a bias in the findings by conducting

independent t-tests comparing participants who completed the entire program and those

participants that did not.

Plan of Data Analysis

The means and standard deviation, of the wait list control group and the treatment

group on the pre-test, post-test and one-month after the prevention program (Acceptance

of Couple Violence Scale and Attitudes Towards Women Scale) was analyzed via a

onetailed t-test. Further, post-hoc analyses were conducted to determine the influence of

certain demographic variables.

CHAPTER FOUR

RESULTS

This chapter describes and summarizes the statistical analyses used to evaluate the

research question and hypothesis established in the previous chapters. For review my

research question is, if a three session gender norms module which is part of the larger

Reaching and Teaching Teens (NDVSAC, 2011) intervention is capable of modifying

gender role beliefs. My hypothesis is: (a) Participants in a 3-session Toxic Scripts

intervention demonstrate greater average stereotypical gender norms attitude change than

participants in a delayed treatment control group, and (b) Differences are maintained over

a one-month post-treatment period.

Subsequent to the data screening process, this chapter reports the descriptive

statistics for demographic variables, data preparation methods, and inferential tests

conducted to test the hypothesis proposed above.

Descriptive Statistics

The preliminary dataset included 23 participants. A total of 14 (66.7%) were

female and 7 (33.3%) were male; two participants did not complete the gender question

on the demographics form. Ages ranged from 14-18 and the average age of participants

was 15.4 years. Table 1 presents an overview of the key demographic and school-related

variables. Missing items for these descriptive statistics are not included and as such,

numbers do not always add to 23 due to not all participants completing the Attitudes

Towards Women Scale (ATWS) at all data points (pre, post and 1-month follow-up).

Table 2 presents a summary of descriptive statistics, arranged by waitlist and

experimental groups.

Table 1. Demographic & School-Level Variables

Demographic Variable

Response Categories

Frequency

Valid Percentage

Gender

Male

Female

33.3

66.7

Grade in School

Ninth

Tenth

Eleventh

Twelve

52.4

28.6

14.3

4.8

Family Composition

Live with both parents

Live with one parent

Live with other family members

28.6

66.7

4.8

Engagement in Afterschool

Activities (can select more

than one)

Sports

Clubs

Council

Other

1 6

43.5

14.3 4.8

28.6

Relationship Status

In a relationship

Have been in a relationship in past

6 months

Have ever been in a relationship

Never been in a relationship

5 2

9 5

23.8

9.5

42.9

23.8

Guardian(s) employment

Full-time Part-time

Out of work < 1 Year

Out of work > 1 Year

Unable to work

3 4

47.6

4.8

19.0

14.3

19.0

Ethnicity

Black / African-American

White / Caucasian

More than one ethnicity

Asian / Pacific Islander

Puerto Rican

Not reported

34.8 17.4

26.1

8.7

4.3

8.7

Experimental / Wait List

Experimental

Wait List

56.5

43.5

Table 2. Demographic & School-Level Variables: Experimental vs. Wait-List Control

Demographic Variable

Response Categories

Experimental

(N, valid %)

Wait List Control

(N, valid %)

Gender

Male

Female

4 (36.4)

7 (63.6)

3 (30.0)

7 (70.0)

Grade in School

Ninth

Tenth

Eleventh

Twelve

3 (23.1)

4 (36.4)

3 (23.1)

1 (7.7)

8 (80.0) 2

(20.0)

0 (0%)

Family Composition

Live with both parents

Live with one parent

Live with other family members

4 (33.3)

7 (58.3)

1 (8.3)

3 (30.0)

7 (70.0)

0 (0.0)

Engagement in Afterschool

Activities (can select more

than one)

Sports

Clubs

Council

Other

7 (53.8)

1 (7.7)

3 (23.1)

3 (30.0)

2 (20.0) 0

(0.0)

3 (30.0)

Relationship Status

In a relationship

Have been in a relationship in past

6 months

Have ever been in a relationship

Never been in a relationship

2 (18.2)

4 (36.4) 3

(27.3)

3 (30.0)

0 (0.0)

5 (50.0) 2

(20.0)

Guardian(s) employment

Full-time Part-time

Out of work < 1 Year

Out of work > 1 Year

Unable to work

6 (54.4) 0

(0.0)

2 (18.2)

2 (18.2) 2

(18.2)

4 (40.0)

1 (10.0)

2 (20.0)

Ethnicity

Black / African-American

White / Caucasian

More than one ethnicity

Asian / Pacific Islander

Puerto Rican

Not reported

5 (38.5)

2 (15.4)

3 (23.1)

0 (0.0)

1 (7.7)

2 (15.4)

3 (30.0)

2 (20.0)

3 (30.0)

2 (20.0)

0 (0.0) 0

(0.0)

Program engagement

Completed program

Did not complete program

52.2

47.8

Program engagement

Completed program

Did not complete program

8 (61.5)

5 (38.5)

3 (30.0)

7 (70.0)

To demonstrate some of the demographic variables visually, graphs are included

below in figures one and two. Figure one presents the relationship status of participants in

the study and figure two presents the ethnicity of participants.

Figure 1. Relationship Status of Participants

Figure 2: Participants’ Ethnicity

relationship

Have

been

relationship

past

months

Have

ever

been

relationship

Never

been

relationship

Relationship

Status

Ethnicity

Black

African‐American

White

Caucasian

than

one

ethnicity

Asian

Pacific

Islander

Puerto

Rican

Not

reported

Scale Preparation

The ATWS consists of 12 Likert scale items, with scores ranging from 1-4. In order to

prepare items for analyses, the scale items were summed and averaged to create a mean

score for each participant. Participants with missing responses to items on the scale were

removed from analyses. This resulted in a lower sample size for scale analyses, which

will be reflected in the inferential tests in the section below. Table 3 and Figure 3 presents

the mean ATWS scores at all three data collection points, arranged by gender.

(Thomas & Hersen, 2011).

Table 3: Attitudes Towards Women Scale Scores at Three Stages, Arranged by Gender

ATWSA Score

Male (M, SD, n)

Female (M, SD)

Pre-Program

2.61 (.26) n=7

3.13 (.29)

n=13

Post Program

2.48 (.36) n=4

3.13 (.57)

n=6

Follow-Up (1 Month)

2.98 (.21)

n=4

3.45 (.25)

n=5

Figure 3. Average ATWS Scores Over Time, Arranged by Gender

0.5

1.5

2.5

3.5

Pre‐Program

Post

Program

Months)

Average

ATWS

Scores

Male

Female

Inferential Tests

First, independent t-tests were conducted to compare ATWS scores of participants

in the experimental and wait-list groups at three data collection points: pre-treatment,

post-treatment, and one month follow up. At the pre-treatment stage, ATWS scores were

nearly identical between the experimental group (M=2.94, SD=.43, n=12) and the waitlist

group (M=2.95, SD=.31, n=10) and there were no significant differences between the two

(t(20)=.33, p=.58). Immediately post-treatment, ATWS scores were slightly higher among

the wait list control group (M=3.22, SD=.79, n=3 ) than the experimental group (M=2.74,

SD=.43, n=8) but this difference was not statistically significant (t(9)=2.24, p=.17).

Finally, an independent t-test was run to test for differences between groups at a follow-

up point one month after the treatment program. At this stage, ATWS scores from the

experimental group (M=3.36, SD=.27, n=3) were slightly higher than the wait list group

(M=3.18, SD=.37, n=6) but this difference was not statistically significant

(t(7)=.60, p=.47). (Thomas & Hersen, 2011).

Table 4. Attitudes Towards Women Scale Scores at Three Stages, for those who

Completed the Program

ATWSA Score

Exp.Group (M, SD)

n=8

Wait-List (M, SD)

n=3

Pre-Program

2.82(.38)

3.11(.42)

Post Program

2.75 (.43)

3.22(.79)

Follow-Up (1 Month)

3.02 (.47)

3.36 (.42)

An independent t-test was conducted to compare ATWS scores of participants in

the experimental and wait-list groups who completed the program (defined as those who

completed the pre-test, post-test and one-month follow-up) at three data collection points:

pre-treatment, post-treatment, and one month follow up. At the pre-treatment stage,

ATWS scores were M=2.82, SD=.38, n=8 for the experimental group and M=3.11,

SD=.42, n=3 for the wait-list group and there were no significant differences between the

two (t(9)= -1.09, p=.94). Immediately post-treatment, ATWS scores for the experimental

group were M=2.75, SD=.43, n=8 and the wait-list control group was M=3.22, SD=.79,

n=3 but this difference was not statistically significant (t(9)= -1.31, p=.18). Finally, an

independent t-test was run to test for differences between groups at a one-month after the

treatment program. At this stage, ATWS scores from the experimental group were

M=3.02, SD=.47, n=8 and the wait list group scores were M=3.36, SD=.42, n=3) but this

difference was not statistically significant (t(9)=-1.09, p=.78). (Thomas & Hersen, 2011).

Post-Hoc Analyses

A one-way analysis of covariance (ANCOVA) was conducted to determine if

there was a difference in scores on the ATWS between the group of students who had

completed the prevention program (n=8) and those who did not (n=3). Completion of the

program was defined as completing the ATWS at all three data points (pre, post and

1month follow-up). The benefit of ANCOVA testing is that it allows for an explanation of

variation along with possible confounding (covariate) variable to be entered into the

equation as a control, permitting the evaluation of change over time. The ANCOVA

examines unexplained variance and attempts to explain it with the covariate in the model.

A total of two ANCOVAs were performed to assess the level of change immediately

following the prevention program as well as any lasting change at one month. (Thomas &

Hersen, 2011).

In the first analysis, the covariate was a participant’s score on the ATWS prior to

the engagement with the prevention program. The dependent variable for the ANCOVA

was the ATWS score immediately after the treatment program. The independent variable

was whether participants were in the experimental group or the waitlist group. Table 4

presents the unadjusted mean scores and the adjusted mean scores (including the

covariate) on the ATWS for both treatment groups.

Table 5. Scores for ATWS Immediately Following Program: Adjusted and Unadjusted

Unadjusted Means

Experimental 2.74

Waitlist 3.22

Adjusted Means

Experimental

Waitlist

2.83

3.00

The ANCOVA was not significant (F (1,9) = .46, p = .52) demonstrating that there

was no significant difference between mean scores on the ATWS between the groups

immediately after the program.

An additional ANCOVA was run to determine if there was a lasting difference in

scores on the ATWS at one month post treatment between those who completed training

(n=2) and those who did not (n=6). In this analysis, the covariate remained a participant’s

score on the ATWS prior to the engagement with the prevention program. The dependent

variable for the ANCOVA was the ATWS score one month later. The independent

variable was whether or not the student had completed the prevention program. Table 5

presents the unadjusted mean scores and the adjusted mean scores (including the

covariate) on the ATWS for both treatment groups.

Table 6: Scores for ATWS at 1-Months Post: Adjusted and Unadjusted

Unadjusted Means

Experimental 3.21

Wait list 3.18

Adjusted Means

Experimental 3.22

Wait list 3.18

The ANCOVA was not significant, F (1,6) = .10, p = .77, demonstrating that there

was no significant difference between mean scores on the ATWS between treatment

groups one month post prevention program.

Finally, a dependent sample t-test was run to determine if there was a difference

between individual mean scores among the experimental group pre-treatment and at 1

month post-treatment. While the mean score at 1-month post was slightly higher

(M=3.17, SD=.48) than pre-treatment (M=2.94, SD=.45), this difference was not

statistically significant (t(9)=-1.80, p=.11).

Additional t-tests were run post hoc to determine what association key

demographic variables (gender, grade, ethnicity), family level factors, and school-level

variables may have with ATWS scores. Gender was proven to be significantly associated

with ATWS scores at pre-test stage (t(18)=4.12, p=.001) with females having a

significantly higher scores (M=3.13, SD=.29) than their male counterparts (M=2.61,

SD=.26). Immediately following the treatment, female scores remained higher (M=3.13,

SD=.57) than males’ (M=2.48, SD=.36) but the difference was not statistically significant

(t(8)=1.99, p=.60).

An independent sample t-test was conducted to determine if family composition

was associated with ATWS scores. However, there was no significant difference in mean

scores between students who resided in single parent (N=14) and those in two parent

homes (N=6) at pre-treatment (t(18)=-.84, p=.42) and post-treatment (t(8)=-.26, p=.80).

An independent sample t-test was conducted to determine if participation in

afterschool activities was associated with ATWS scores. There was a significant

difference in mean scores between students who engaged in afterschool activities (N=7)

and those who did not (N=1) at pre-treatment (t(6)=2.54, p=.05) but there was no

significant difference post-treatment (t(7)=-.03, p=.98). This finding is tentative given the

small comparison group (N=1).

The potential relationship between age and ATWS scores was explored using a

non-parametric Spearman’s correlation test. Spearman’s tests were selected for this

analysis due to the small sample size and the non-normative distribution of the ATWS

variable (Thomas & Hersen, 2011). The analysis determined that there was no significant

correlation between age and ATWS scores pre-treatment (r=-.01, n=20, p=.98) but a

slightly significant correlation emerged immediately following the treatment (r=-.65,

n=10, p=.04

CHAPTER FIVE

DISCUSSION

This chapter will discuss the implications of the results presented in the previous

chapter (Chapter 4: Results). First, the findings of the main and post-hoc analyses will be

discussed along with the possible explanations of the results and how they are related to

the existing literature. Next the study’s limitations will be discussed along with

implications for future prevention work and research.

Main Findings

This study’s hypothesis was that participants in a 3-session Toxic Scripts

intervention would demonstrate greater average stereotypical gender norms attitude

change than participants in a delayed treatment control group and that these differences

would be maintained over a one-month post-treatment period. Independent t-tests were

conducted to compare ATWS scores of participants in the experimental and wait-list

groups at three data collection points: pre-treatment, post-treatment, and one month

follow up. Further, ANCOVA’s were conducted to allow for an explanation of variation

along with possible confounding (covariate) variable to be entered into the equation.

There were mixed findings in this study, some significant and some non-significant.

At the pre-treatment stage, ATWS scores were nearly identical between the

experimental group and the wait-list group and there were no significant differences

between the two. Immediately post-treatment, ATWS scores were slightly higher among

the wait list control group than the experimental group but this difference was not

statistically significant. Finally, at one-month follow-up ATWS scores from the

experimental group were slightly higher than the wait list group but this difference was

not statistically significant. Both groups started out with essentially the same scores on

but failed to make statistically significant changes in their ATWS scores after going

through the prevention program. These results are contrary to the successful results from

other violence prevention programs that have a focus on gender norms Foshee , Bauman,

et al., 2006;Foshee et al., 1998, 2000, 2004, 2005; Weisz & Black, 2001;Wolfe et al.,

2003; Jaycox et al., 2006, & Barker et al, 2010)

Post-Hoc Analyses

Further analyses were conducted to allow for an explanation of variation along

with possible confounding (covariate) variable to be entered into the equation. In the first

analysis, the covariate was a participant’s score on the ATWS prior to the engagement

with the prevention program. The dependent variable for the ANCOVA was the ATWS

score immediately after the treatment program. The independent variable was whether

participants were in the experimental group or the waitlist group. The ANCOVA was not

significant demonstrating that there was no significant difference between mean scores on

the ATWS between the groups immediately after the program.

An additional ANCOVA was run to determine if there was a lasting difference in

scores on the ATWS at one-month post treatment between those who completed training

and those who did not. In this analysis, the covariate remained a participant’s score on the

ATWS prior to the engagement with the prevention program. The dependent variable for

the ANCOVA was the ATWS score one month later. The independent variable was

whether or not the student had completed the prevention program. The ANCOVA was

not significant, demonstrating that there was no significant difference between mean

scores on the ATWS between treatment groups one-month post prevention program. It

appears that subtracting the variance attributable to participants’ ATWS score at pre-test

and whether or not they completed the program or not did not lead to statistically

significant findings. Again, this finding is inconsistent with the successful results from

other violence prevention programs that have a focus on gender norms (Foshee , Bauman,

et al., 2006;Foshee et al., 1998, 2000, 2004, 2005; Weisz & Black, 2001;Wolfe et al.,

2003; Jaycox et al., 2006, & Barker et al, 2010)

Key Demographic Variables

Gender

Post-hoc analyses were run to determine what association key demographic

variables (gender, grade, parents), family level factors, and school-level variables may

have with ATWS scores. Gender was proven to be significantly associated with ATWS

scores at pre-test stage with females having significantly higher scores than their male

counterparts. Immediately following the treatment, female scores remained higher than

males’ but the difference was not statistically significant. Girls began the program with

less stereotypical gender norm attitudes than boys. Weisz and Black (2001) in their study

using the Reaching and Teaching Teens Curriculum found that girls in their study did had

less violence-supportive attitudes than their male counterparts. This finding was

consistent with other studies that also found that girls had less accepting attitudes of

violence than boys (Jaffee et al.,1992 & Krajewski et al., 1996). In contrast, Wolfe et al.

(2009) found in their study, reductions in dating violence for boys but not for girls.

Further, Foshee et al. (2009) and Taylor, Mumford and Stein (2014) found the Safe Dates

and Shifting Boundaries programs respectively found both programs to be equally

effective for boys and girls in reducing dating violence behavior. The differing results for

the effectiveness of dating violence prevention programs for boys and girls may be due to

the program content and what exactly the authors were measuring. There is little

consistency in the measures used for attitudes and behaviors in the dating violence

literature, which makes it difficult to have a consistent finding among researchers.

Afterschool Activities

Participation in afterschool activities was associated with ATWS scores. There

was a significant difference in mean scores between students who engaged in afterschool

activities (N=7) and those who did not (N=1) at pre-treatment but there was no significant

difference post-treatment. This is a tentative finding due to the comparison group’s

sample size of 1. Such a small comparison group makes it difficult to make a decisive

statement regarding such results (Thomas & Hersen, 2011). The studies discussed in

chapter 2 of this proposal did not take into consideration the effects of afterschool

activities or extracurricular activities on adolescent’s attitudes or behavior

(Wolfe et. al, 2003; Foshee et al., 1998; Foshee et al., 2005; Jaycox et. al.2006 & Weisz

& Black, 2001). Other studies beyond the dating violence literature have indicated that

afterschool activities or extracurricular activities are associated with improved academic

outcomes (higher attendance rates, reduced suspensions, higher grades, and lower

dropout rates); “better social/emotional outcomes (decreased behavioral problems,

improved social and communication skills, increased self-confidence and self-efficacy

and lower levels of depression and anxiety) (Harvard Family Research Project, 2008,

p.5);” better prevention outcomes (reduction in drug and alcohol use, decreases in violent

behavior and increased knowledge of safe sex) and better health and wellness outcomes

(increased physical activity, healthier eating and improved body image). Therefore, it is

possible with a larger sample size, one would obtain a greater difference on the ATWS

between those who engaged in afterschool activities and those who did not engage in

such activities. It is possible that those who engaged in afterschool activities would have

attitudes less accepting of stereotypical gender norms (Harvard Family Research Project,

2008).

Age/Grade Level

The potential relationship between age and ATWS scores was explored using a

non-parametric Spearman’s correlation test. Spearman’s tests were selected for this

analysis due to the small sample size and the non-normative distribution of the ATWS

variable. The analysis determined that there was no significant correlation between age

and ATWS scores pre-treatment but a slightly significant correlation emerged

immediately following the treatment. This is an interesting finding which is consistent

with the literature that shows that dating violence peaks around age 16 or about 10th grade

and subsequently declines (Foshee et al. 2009; Reyes et al. 2011). With a decline in dating

violence, as one gets older, it is also likely that they subscribe to less stereotypical gender

norms. This also follows the developmental trajectory in which early adolescence is

marked by sex segregated play and more rigid gender norms attitudes in which younger

adolescents rely more on gender information than individual information while older

adolescents are able to take into account exceptions to these stereotypes (Biernat, 1991).

As adolescents developmentally progress they become less rigid in their gender norm

attitudes and are able to demonstrate a flexible application of gender stereotypes (Biernat,

1991).

Limitations

Despite the importance of prevention programs, there are many challenges to

implementation, some of which were relevant for the current project. Some of these

hurdles to implementation include: small sample size, difficulty with recruitment,

attrition, dosage, and possible ineffectiveness of the program. These obstacles will be

discussed in the following sections.

Small Sample Size

In this study this researcher was only able to recruit 23 participants in total for

both conditions. There were 13 participants in the experimental group and 10 in the

waitlist control group. Of these participants, only 8 participants in the experimental group

completed the program (completed assessments at all three data points) and only 5

participants completed the program in the wait-list control group. This left only 13

participants ATWS scales for analysis. This was far short of this researcher’s desire for

60 or more particpants to obtain a medium effect size (Ting, 2009 & Thomas & Hersen,

2011). The difficulties of recruiting participants and minimizing attrition will be

discussed in the following paragraphs. These difficulties directly affect the sample size,

which consequently may have affected the ability to find statistically significant

differences between the two groups.

Participants were not able to be matched in pairs based on age or grade level and

then randomly assigned to either the treatment or wait-list control group due to the low

sample size. Due to the low level of participation, this researcher had to place participants

in the group that best fit their schedules. Therefore, this researcher was not able control

for individual differences between participants which may have effected the outcomes

(Thomas & Hersen, 2011).

The difficulties with a small sample are widely found in the dating violence

prevention literature (Foshee & McNaughton Reyes, 2009). Many studies are

underpowered due to the difficulty of recruiting and retaining participants for this type of

research. The evaluation of prevention programs is necessary to determine program

effectiveness. “Otherwise, practitioners may assume that a program is effective on the

basis of anecdotal or case study evidence. As evaluation has become more common, the

results indicate that many programs that are anecdotally believed to be successful may

actually not be effective (Nation et al., 2003 p.458).”

Recruitment

In this study there was a lack of adequate participation. This researcher was only

able to recruit 23 participants for the study. This was short of the 60 participant

recruitment goal. Other studies conducted on youth prevention programming have

highlighted the difficulty of recruitment. “Rates for youth-only programs targeting risk

behaviors, such as smoking or drug use, have been reported at 50% acceptance rates, with

short- term retention in intervention studies at 50 to 80% (Hooven, Walsh, Willgerodt &

Salazar, 2011 p. 138). Some of the reasons for the low acceptance or recruitment rates,

which will be discussed in the following paragraphs, are a lack of support from schools

and community agencies, competing after-school activities, lack of clear communication

and length of recruitment.

Lack of Support from School and Community Agencies

A recruiting challenge found in this study, which is also echoed in the prevention

literature, is the lack of support from schools and the community (Hayes, Chapple &

Ramirez, 2014, Whitaker & Lutzker, 2009 &Weisz & Black, 2009). Schools and

community agencies are limited on time and finances. The focus of the schools and

agencies is on academic performance and current problem behaviors. This attention is

given to the detriment of attending to social and emotional needs and prevention. Many

agencies fail to make the connection between emotional health and academic success

(Weir, 2012). Although, Nebraska schools are required by the Lindsay Ann Burke Act

(Nebraska Department of Education 2010) to provide dating violence education, this

usually occurs in the format of a one-day informative session by a staff of the local

domestic violence shelter and is not specific to prevention.

Even with community and school support, researchers have found it difficult to

recruit participants for dating violence prevention programs. Although, researchers for the

Ending Violence prevention program had support from many of the local schools and

recruited 55 classes in 10 schools for participation in the program, the researchers lost a

total of 3 schools from the program due to logistical problems (Jaycox et al., 2006). The

evaluation on the Reaching and Teaching Teens curriculum (Weisz & Black, 2003) had

support from a local school that was affiliated with the researchers’ university. Even with

this support, the researchers were only able to recruit 46 students in the intervention

group and 20 students in the comparison group of the 250 students who were eligible to

participate (Weisz & Black, 2003). Unfortunately, for this study, this researcher was not

able to obtain the support from the school system to implement the Reaching and

Teaching Teens Curriculum. This researcher had to recruit community centers and

organizations that were not school-affiliated. The school system in Lincoln, Nebraska has

a rigorous review board designed for deciding which research projects should be allowed

into the schools. Research projects outside of teacher and student academic education

have limited success in entering the school system. The Lighthouse was the only

community organization, of the many this researcher met with, that was excited and

willing to bring a dating violence prevention program into their center. A major theme in

the dating violence prevention literature is the importance of building a relationship with

the agency for a significant period of time and enlisting their input on the curriculum

(Hayes, Chapple & Ramirez, 2014; Vera & Speight, 2003; Whitaker & Lutzker, 2009,

Weisz & Black, 2009, Vera, 2013). Although, this researcher met with staff from the

Lighthouse numerous times prior to and during the prevention program, there was not a

significant relationship with the agency prior to implementing the prevention program. In

the future, this researcher and other researchers should spend more time developing the

relationship with the agency they will conduct prevention work in before the program

begins.

Hayes, Chapple & Ramirez (2014) used specific effective strategies that would

provide more support from agencies. These included attending functions at the school or

agency, having meetings with staff to educate them about the program, providing

informational sessions with incentives and enlisting staff to provide suggestions and

recommendations. Some similar recruitment practices were used in this study. In

particular, this researcher met with staff numerous times and with the Lighthouse

Director. This researcher enlisted their insight on the curriculum. These efforts may have

helped with some of the participation recruitment, but not to the level expected. In the

future, this researcher will use the tactics described above to enlist more participants.

Competing After-School Activities

Another recruiting challenge found in this study, which is also found in the

prevention literature, was difficulty with competing after-school activities (Hayes,

Chapple & Ramirez, 2014, Whitaker & Lutzker, 2009 &Weisz & Black, 2009). One study

by Hayes, Chapple and Ramirez (2014) discussed the difficulties with recruiting

adolescents due to many competing after-school activities. These after-school activities

may not need participants to receive parental consent or fill out assessments prior to

participation (Hayes, Chapple & Ramirez, 2014). Some popular after-school activities

include athletics, clubs, jobs, and leisure activities. Besides not needing parental consent

to join these activities, for many youth they are more appealing than attending prevention

programs. Attendance has shown to be directly correlated with incentives (Hayes,

Chapple & Ramirez, 2014; Weisz & Black, 2009 & Jaycox et al., 2006). One study found

that when they offered pizza the attendance was higher and when they stopped offering

pizza the attendance went down. Further, when students were more likely to receive an

incentive (the odds are in their favor) then attendance was higher. In particular, Hayes,

Chapple and Ramirez (2014) compared size of schools, the average attendance rate for

the two largest schools was 51%; whereas, for the two smallest schools it was 70%.

Weisz & Black (2003) offered gift cards to youth who participated which increased

participation rates. In this study incentives were offered. All participants were entered

into a drawing for a chance to win an IPod. Further, candy bars were given to those who

brought back consents and participants were provided with food and beverages at each

prevention session. These incentive efforts may have helped in some participation

recruitment, but not to the level expected.

Hooven, Walsh, Willgerodt and Salazar (2011) suggested offering not only

incentives for participants but to also provide incentives for the community agency staff.

These incentives could improve the staff’s view of the program and may make them more

likely to promote the program to the youth. In the future, this researcher will use more

incentives for youth as well as provide incentives to the agency staff, to increase

participation rates.

Communication Difficulties

Another recruiting challenge found in this study and consistent with the dating

violence prevention literature, was difficulty in communication with agency staff (Hayes,

Chapple & Ramirez, 2014; Whitaker & Lutzker, 2009; Weisz & Black, 2009;Wolfe et. al,

2003; Foshee et al., 1998; Foshee et al., 2005; Jaycox et. al.2006 & Weisz & Black,

2001). Jaycox et al. (2006) had difficulties with their prevention program due to high

staff turnover and miscommunication between researchers and staff. The staff liaison for

this study often miscommunicated dates and times of when the prevention program could

begin and days to recruit students. This possibly led to participants who were initially

interested in the program but may have lost interest. This loss of interest could be due to

the continual rescheduling of the prevention program to a later start date by staff.

Research has shown that once a participant has consented to participate in a program, the

program should begin not long after. This is due to a loss of interest and other competing

activities and life events that may occur from the time a participant consents and actually

participates (Hayes, Chapple & Ramirez, 2014& Weisz & Black, 2009). A way to

improve communication between researchers and agency staff is the researcher to provide

all staff at the agency with the same information by many different mediums; email,

phone calls and printed information. Meetings with staff, informational brochures and

staff bulletins were also used by researchers to improve communication (Hooven, Walsh,

Willgerodt & Salazar, 2011 & Jaycox et al, 2006). In the future, this researcher will use

these aforementioned effective communication practices to improve researcher and staff

communication, which in turn may influence participation rates.

Length of Recruitment

Lastly, another recruiting challenge found in this study, which is also echoed in

the prevention literature, is the amount of time spent recruiting (Hayes, Chapple &

Ramirez, 2014; Whitaker & Lutzker, 2009; Weisz & Black, 2009;Hooven, Walsh,

Willgerodt & Salazar, 2011 & Jaycox et al, 2006). Although, researchers talk about the

importance, there has not been consensus among researchers concerning the appropriate

amount of time to recruit.

In this study, the recruitment time may have been too lengthy. This researcher’s

recruitment time was lengthy due to needing enough participants to obtain significant

results. Due to this, many of the consents and assents were forgotten or lost which

resulted in a loss of resources and a lengthier time to complete the recruitment. Although,

the ideal amount of recruiting time is unknown, research has shown that the intervention

should begin not long after participants sign consents due to loss of interest or other life

events happening (Hayes, Chapple & Ramirez, 2014& Weisz & Black, 2009).

Attrition

There were a large percentage of participants in this study who started the

prevention program but did not continue on to complete the program. Although both

groups had high attrition rates, the wait-list control group had a slightly higher attrition

rate than the experimental group. This difference was not statistically significant. The

experimental group had 61.5% of the participants complete the program while 38.5% did

not complete the program (complete ATWS at all three data points). In the control group,

only 30% of the wait-list control group completed the program while 70% did not

complete the program. The attrition rate in the experimental group is consistent with the

dating violence prevention literature, which suggests that 40–50% attrition is

commonplace in longitudinal studies (Horton & Lipsitz, 2001 and Foshee, 2005). The

high rate of attrition could be due to the length of the program and a lack of interest by

participants.

Attrition with Increased Length of Program

Research shows that attrition rates increase with the length of the program

(Horton & Lipsitz, 2001 and Foshee, 2005). In particular, The Safe Dates program had

“attrition from baseline by wave 3 was only 12.0%, but by wave 5 was about 50%

(Foshee et al., 2005, p.249).” In the Youth Relationships Project prevention program, the

attrition rate was 50% (Wolfe et al., 2003). Wolfe et al. (2003) reported that retention

rates of participants in the program continued to decrease as the program continued.

Particularly, “The percentage of the final sample that had two or more follow-up

assessments was 2 assessments (94%), 3 (81%), 4 (75%), 5 (65%), 6 (50%); less than half

the sample had 7 or more follow-up assessments (Wolfe et al., 2003,p. 281). The Ending

Violence program (Jaycox et al.2006) also had difficulty with attrition. Jaycox et al.

(2006) reported that 2540 of 3800 enrolled students completed the assessments. In

particular, “67% participation rate overall; 71% intervention, 62% delayed control and of

these, 13% missed the pretest survey, 12% missed the posttest survey, and 24% missed

the follow-up survey (Jaycox et al.2006 p. 697).” This study was designed to be shortterm

to be amenable to implementation. A shorter program would seem to reduce attrition

rates. This may not have been the case in this study because of the inconsistent spacing of

the sessions. The study was designed to be three sessions conducted in three consecutive

weeks. Due to the Lighthouse’s schedule and the staff liaison for this study often

miscommunicating dates and times of when the prevention program, the sessions were

spaced sporadically and not in consecutive weeks. Research has shown that only a short

time should elapse between sessions to reduce attrition (Hayes, Chapple & Ramirez,

2014& Weisz & Black, 2009). Researchers suggest that a way to ameliorate this problem

is to have back-up days built into the curriculum to reduce the time between sessions if a

session is rescheduled (Hayes, Chapple & Ramirez, 2014& Weisz & Black, 2009). In the

future, this researcher will have back-up days built into the schedule to reduce attrition.

Participants’ Interest

Participation in this program was voluntary and competed with some popular

after-school activities including athletics, clubs, jobs, and leisure activities. Many of the

participants may have found these activities more appealing than attending a dating

violence prevention program. Discussing gender norms may have been uncomfortable for

many of the participants whose first time discussing these issues may have been in this

program.

To reduce attrition rates, the prevention literature has shown that researchers need

to have a strong buy-in from participants, strong agency support and provide incentives

for the participants (Hayes, Chapple & Ramirez, 2014; Weisz & Black, 2009 & Jaycox et

al., 2006). To increase participant buy-in or to make the program more relevant and

valuable to participants Vera and Speight (2003) suggested seeking participants’ input on

the program and making the program culturally relevant to the participants. If participants

have a stake in an intervention then they are less likely to drop out before the completion

of the program (Vera & Speight, 2003). Further, having constant contact with agencies,

participants, and participant’s guardians are crucial to reducing attrition rates. In

particular, having increased visibility at agency events and after-school activities allows

for staff, participants and participants guardians to have an understanding of the

importance of the program. (Hayes, Chapple & Ramirez, 2014& Weisz & Black, 2009).

Increasing the use of incentives has shown to decrease attrition rates. Youth are more

likely to attend prevention sessions in which they are given some sort of reward (Hayes,

Chapple & Ramirez, 2014& Weisz & Black, 2009). Although, incentives were used in

this study, the incentives may need to be more desirable to the participants. In the future,

this researcher will utilize the above techniques to reduce attrition rates.

Dosage

Adolescent Dating violence prevention programs have varied in dosage or

program intensity. Nation, et al.’s (2003) review of effective prevention approaches to

adolescent risky behavior defines sufficient dosage as, “the need for participants to be

exposed to enough of the program for it to have an effect. Dosage, or program intensity,

may be measured in quantity and quality of contact hours. Aspects of dosage include the

session length, number of sessions, spacing of sessions, and the duration of the total

program (p. 452).” The recommended dosage for the entire Reaching and Teaching Teens

curriculum is 12 sessions. The recommended dosage for the Toxic Script unit varies from

one to three sessions. The only evaluation, besides this study, was on the entire program.

Results indicated that knowledge of sexual assault and dating violence increased at the

end of the program and was maintained at six-month follow-up, as well as an

improvement in attitudes of sexual assault and dating violence (decreased acceptance of

dating violence and sexual assault) (Weisz & Black, 2001). The authors conducted the

program in twelve, one and one half hour sessions over a six-week time period (Weisz &

Black, 2001). The length of the program, spacing of sessions and booster-sessions will be

discussed below in regards to the appropriate dosage.

Length

There is little empirical research about the minimum number of sessions a

prevention program can be to see results. Research does show the ineffectiveness of a

one-session format for lasting change (Foshee & McNaughton-Reyes, 2009). This

researcher discussed in chapter two about the evidence for short-term prevention

programs. In particular, Jaycox and colleagues (2006) found that a three sessions of

treatment showed changes in altering norms and behaviors regarding dating violence.

Lavoie, Vezina, Piche and Boivin (1995) evaluated the effects of short-term (i.e., 2

sessions) versus long-term (i.e., 4 sessions) dating violence prevention programs. The

results showed no differences between the long and short-term programs. This researcher

went with a three- session format due to the evidence in the literature and feasibility of

short-term prevention programs (time limitations in community agencies and schools and

financial limitations). The two programs referenced above measured a change in dating

violence norms rather than gender norms. The dosage or length of the program may not

have been what is needed for a change in gender norms. Gender norm change may take

longer. A three-session format used in the Toxic Script lesson of the Reaching and

Teaching Teens curriculum may have not been a sufficient length of time to make

changes in gender norms.

The goals of the Toxic Script unit in the Reaching and Teaching Teens curriculum

are to increase awareness of gender norms, knowledge of gender norms and decrease

attitudes accepting of stereotypical gender norms and increase gender flexibility. To arrive

at these goals may take longer than three sessions and a greater length of time for sessions

because gender norms are embedded in our society and organize our everyday behavior.

Messages of stereotypical gender norms are communicated everywhere in youth’s lives

including parents, school, media and peers (Flay, 2000; Flay & Allred, 2001). Further,

adolescents are at a developmental time when their gender norms are more rigid and the

differences between the genders are highlighted. The deep entrenchment of stereotypical

gender norms and the developmental stage of adolescence may require lengthier

prevention programs to produce a change. It is possible that the unit Toxic Scripts in the

Reaching and Teaching Teens Program may be effective, but was not conducted for a

long enough period of time (Flay, 2000; Flay & Allred, 2001 & Foshee & McNaughton

Reyes, 2009).

Spacing of Sessions

The study conducted by this researcher was designed to have the prevention

sessions once a week for three consecutive weeks. Due to scheduling conflicts with the

organization, this was not achieved. The prevention sessions were spaced several weeks

apart. The Lighthouse’s scheduling conflicts were often last minute and this researcher

arrived at the Lighthouse for the planned dates of the prevention program. This researcher

was then told that there was a scheduling conflict and the participants would not be able

to attend. This researcher would then have to reschedule the session and inform the youth

through staff, which created a greater length of time between sessions than was intended.

The inconsistent delivery of the message may have impacted the effectiveness of the

intervention. In the future, this researcher will communicate more effectively to the

agency staff the importance of having the prevention program disseminated in a

consistent manner. Further, this researcher will utilize what other researchers have and

have back-up days built into the schedule in the case of such a last minute rescheduling

(Foshee et al., 2005 & Jaycox et. al., 2006) Having back-up days built into the schedule

will help lessen the amount of time between sessions (e.g. have back up day built into the

schedule which are the same week of the missed prevention session).

Booster-Session

Research shows that a majority of the effects in prevention programs are lost over

time. (Nation et al., 2003). A booster-session is recommended that reiterates and

summarizes what was already learned to prevent this decline (Nation, et al, 2003). Foshee

et al. (2005) utilized a booster-session for the Safe Dates program and found lasting

effects. In the future, this researcher, in addition to evaluating the results of the prevention

program over time, will also incorporate a booster-session into this curriculum in order to

improve the chances of maintaining lasting effects

Ineffectiveness

While power issues may explain the lack of significant results, it is also a

possibility that the curriculum was not effective. The Reaching and Teaching Teens

curriculum has not been extensively evaluated. To date, besides this study, there is only

one evaluation on this program (Weisz and Black, 2003). Even though this program has

not been systematically evaluated it is widely used throughout the state of Nebraska

(Zinke, 2013). Future studies should be conducted on this curriculum to ensure that it is

effective for preventing dating violence in adolescents.

There is a large amount of evidence that shows the role that gender norms play on

adolescent risky behavior and the effectiveness of addressing gender norms in prevention

programs (Raithal, 2003; Graine 2009; Schulte, Ramo & Brown, 2009; Schulte, Ramo &

Brown, 2009; Barker et. al., 2010; Pulwerwitz & Barker, 2008; Pulerwitz, Michaelis

Verma & Weiss, 2010; Flay, 2000; Flay & Allred, 2001; Eisler et al., 2000; Parrot &

Zeichner, 2003; Rando, Rogers, & Brittan-Powell, 1998; Robinson & Schwartz, 2004;

Schwartz, Merta, Waldo, & Bloom-Langell, 1998; Reed, Raj, Miller & Silverman, 2010;

Foshee , Bauman, et al’s, 2006;Foshee et al., 1998, 2000, 2004). A few reasons this

study’s theory may have been on track, but the project failed to find signficant findings:

the measure used, this researcher’s fidelity to the program, and using a lesson instead of

the entire prevention program.

Measure

The measure this researcher chose to use in this study was Attitudes Towards

Women Scale-Adolescent (ATWS)(Galambos, Petersen, Richards, & Gitelson, 1985;

Dahlberg, L.L., Toal, S.B., Swahn, M., & Behrens, C.B., 2005). This instrument was

chosen for the following reasons; its ability to measure stereotypical gender norm

attitudes, its appropriateness of use with adolescents and its appropriateness of use with

ethnic minorities (Galambos, Petersen, Richards, & Gitelson, 1985; Dahlberg, L.L., Toal,

S.B., Swahn, M., & Behrens, C.B., 2005). There is a possibility that the ATWS could

have not accurately measured gender norm attitudes for this sample. In particular, the

average score on the ATWS scale for the two groups was relatively high and the average

score for girls was higher than boys. Some research suggests that girls are more sensitive

to questions on the measure than boys and that the scale is more effective in use with

early to middle adolescents because a ceiling effect is found when older adolescents

(particularly girls) are tested (Galambos, 2012).

Although, the Reaching and Teaching Teens curriculum addressed both male and

female gender norms, this researcher only measured attitudes towards women. In the

future, this researcher and other researchers should incorporate scales that measure

attitudes towards female and male gender norms. Two other scales would be beneficial to

use in the future, the Male Role Attitudes Scale (MRAS; Pleck, Sonenstein, & Ku,

1994) and the The Adolescent Masculinity Ideology in Relationships Scale (AMIRS;

Chu, Porche & Tolman, 2005). The MRAS is an 8-item scale that measures masculinity

in terms of attitudes toward male roles and measures, “the perceived importance of men

fulfilling traditional masculinity standards (Thompson & Pleck, 1995, 146).” The AMIRS

was designed to “incorporate the fact that it is often through and within the contexts of

interpersonal relationships that masculine norms are introduced, reinforced, incorporated,

and perpetuated in ways that become personally meaningful and directly consequential to

adolescent boys (Chu, Porche & Tolman, 2005, p. 97).” Both scales have acceptable

reliability and validity and can be used with a diverse population.

Recent research in adolescent dating violence prevention has highlighted the need

for new measures to more accurately measure gender role attitudes (Reyes, Foshee,

Niolon, Reidy, Hall, 2016). Reyes et al., (2016) critique the existing gender role attitudes

measures as outdated and lack applicability to youth. Further, they indicate that, “future

research include measures that specifically assess interpersonal aspects of gender roles

including adversarial relationship beliefs (e.g., the expectations that relationships involve

exploitation, manipulation and deceit) and tap into the ways in which gendered power

relations manifest in dating relationships (as opposed to marital relationships) may be more

directly related to dating violence and relevant to today’s youth than measures focused

exclusively on men and women’s roles in society (Reyes, Foshee, Niolon, Reidy,

Hall, 2016, p. 357).”

Fidelity to the Program

This researcher did not have the benefit of having an independent observer

complete a fidelity measure to measure how well this researcher adhered to the program.

This fidelity measure could have been used to analyze the results and see if this

researcher’s level of adherence to the lesson affected the participants’ outcomes (e.g.

lower scores on fidelity checklist could be associated with smaller changes in gender

norm attitudes). Therefore, this researcher only has a subjective view of how closer the

program was followed. Although this researcher views that the program was closely

adhered to, there was no objective measure completed. Consequently, the results may

have been affected by a possible low adherence to the program

Other researchers in the dating violence prevention literature have measured

facilitator fidelity to the program. Jaycox et al. (2006) assessed fidelity to the Ending

Violence curriculum through an expert who observed classes randomly and assessed

fidelity as well as having the presenter’s rate their own performance on fidelity. Weisz &

Black (2001) took another approach and had facilitators meet on a weekly basis to discuss

their overall fidelity to the curriculum and ways to improve their adherence for the next

prevention session. Wolfe and colleagues (2003) assessed for fidelity to the Youth

Relationship Project curriculum by videotaping all sessions and had raters score the

facilitators adherence to the protocol. Although all of the above assessed for fidelity, and

reported a strong adherence to the program, none of the researchers reported how fidelity

to the protocol may have influenced outcomes. In the future, this researcher will utilize a

fidelity protocol in order to maintain adherence to the program.

Lessons Versus Entire Program

This researcher chose to use a segment of the Reaching and Teaching Teens

curriculum instead of the entire curriculum. The reasons for this were outlined in earlier

chapters and included; a need for short term prevention programs, this study’s focus on

changing stereotypical gender norms and the frequency of use by facilitators of only

segments of the curriculum versus the entire curriculum. An informal survey indicated

that only 9% of dating violence prevention facilitators throughout the state of Nebraska

used the entire Reaching and Teaching Teens curriculum (Zinke, 2013). The majority of

facilitators use parts of the curriculum due to time constraints and financial limitations.

In the dating violence prevention literature there is some push for modulation of

curriculums or “dividing the curriculum into constituent parts (Kerin, Volz, Moeddel,

2010 p.675).” The call for modulation of prevention programs would enable facilitators

the flexibility to adapt the program to their audience while still maintaining the essential

parts or ingredients of a program (Lock, 2004 & Kerin, Volz, Moeddel , 2010). This

flexibility is beneficial for facilitators but is not ideal for researching the effectiveness of

a program. In order to evaluate the effectiveness of a program, researchers need

facilitators to adhere strictly to the curriculum or maintain a certain level of fidelity to the

program to ensure an accurate measurement.

Although there is a push for modulation of curriculums, there is lack of research

outlining the effectiveness of this approach. The lack of significant findings in this study

may be due to this researcher’s use of a lesson in the program instead of using the entire

program. In the future, this researcher will utilize the entire program to evaluate its

effectiveness. There has not been much research on using components of a dating

violence prevention program and its effectiveness. Further research, should be done on

modularization due to the reality that facilitators rarely use the entire program due to time

constraints and financial limitations.

Differences between Ethnicities in Gender Norms

Of the nearly 1.7 million individuals in Nebraska, 90% are White, 5.5% are

Hispanic, 4% are African American, 2% Asian and 1% are Native American (Nebraska

Department of Health and Human Services, 2010). This researcher attempted to enlist a

similar demographic makeup for the sample in this study. Lincoln and Omaha are the two

largest cities in Nebraska and tend to be more diverse than the majority of the state at

86% White, 6% Hispanic, 4% African American, 4% Asian, 0.8% Native American and

those reporting two or more races is 3% (United States Census, 2014). The Lighthouse

is located in one of the most diverse neighborhoods in the city where the ethnic makeup

is 78% White, 7% African American, 2% Native American, 4% Asian, 5% Hispanic and

5% reporting two or more races (Lincoln Department of Urban Development, 2015). The

majority of participants in this study identified as either African American and more than

one ethnicity. The Reaching and Teaching Teens program was constructed to be culturally

sensitive and applicable to a diverse range of participants. The one research study

conducted on this curriculum was with a sample of majority of African-American

participants and the results showed improvement in knowledge of sexual assault and

dating violence as well as decreased acceptance of dating violence and sexual assault

(Weisz & Black, 2001). The authors didn’t measure gender norm attitudes. Research on

gender norm attitudes has indicated that there are differences among ethnicities (Kane,

2000). Studies show that African-American’s tend to have more gender equitable

attitudes than Whites (Gondolf& Williams, 2001 in Foshee book). Hispanics and

AsianAmericans tend to have more stereotypical gender norm attitudes than Whites and

African-Americans (Kane, 2000 & Gondolf & Williams, 2001 in Foshee book). While

differences exist between ethnicities, there are variations within each group. Individuals

who are more acculturated tend to have less stereotypical gender norm attitudes than

individuals who are more acculturated. A lack of research exists about many other

ethnicities gender norm attitudes. More research is needed on how gender norm attitudes

manifest in these cultures. Further, data from the United States Census Beaue (2014)

indicate that 2.5% identify as two or more ethnicity and that is projected to increase to

6.5% by 2060. In the future researchers should also study how gender norms attitudes are

displayed in individuals who identify as having more than one ethnicity.

The program may not have been effective because it may not have fully address

how different cultures view gender norms and therefore participants may felt the program

was not relevant to them. Further, the majority of participants were African-American and

they may have less stereotypical gender norms and therefore not have much change to

make. In the future, this researcher and other researchers should make sure to address

differences in gender norms among different ethnicities to ensure the program is relevant

for all participants.

Community versus School-Based Prevention Programs

This researcher implemented a community based prevention program due to the

difficulty in implementing prevention programs in schools. This is due to pressure placed

on schools to include as much academic work as possible and to minimize any time

considered to be non-academic. Researchers often find it difficult to persuade school

administrators and teachers to allow them more than one session to provide programming

to students. (Weisz & Black, 2009). After implementing a community-based intervention,

this researcher recognizes that challenges exist in both community and school domains.

Implementing a community-based dating violence intervention program was extremely

difficult due to recruitment challenges, attrition and staff communication. Community

agencies, like schools, have their own rules and regulations that serve its purpose for the

agency or school but often are an impediment to implementing and evaluating a

prevention program. Schools are a good vehicle for prevention due to having a captive

audience which leads to higher recruitment rates and lower attrition rates. Although, there

are difficulties entering into schools, the problems once a researcher is able to get into a

school are minimized. While it may be easier to get into community agencies, the

problems are vast once implementing the program. Schools are a good vehicle for

prevention programs in order to minimize attrition, maximize sample size and the ability

to systematically evaluate programs (Hayes, Chapple & Ramirez, 2014, Whitaker &

Lutzker, 2009 &Weisz & Black, 2009).

Conclusion

Adolescent dating violence is widespread and has many negative repercussions

(Lewis & Fremont, 2000; Foshee & McNaughton Reyes, 2009; Exner-Cortens,

Eckenrode & Rothman, 2013; & NCDVA, 2013). Nebraska youth are especially at risk

for experiencing dating violence (CDC, 2012). There is a clear need to have effective

dating violence prevention programs. While, Reaching and Teaching Teens, is used

widely throughout the state, there has only been one study conducted on its effectiveness

(Weisz & Black, 2001). Further, there have been studies showing the effectiveness of

dating violence prevention programs that provide education on gender norms (Foshee,

Bauman, et al., 2006; Foshee et al., 1998, 2000, 2004; Wolfe et al., 2003; Jaycox et al.,

2006). In this study, this researcher sought to see if a dating violence prevention program,

Reaching and Teaching Teens, aimed at changing gender norms was effective for

adolescents. This author did not find a statistically significant effect in gender norms

attitudes. There are multiple reasons this may have been the case; small sample size,

difficulty with recruitment, attrition, dosage, measurement, gender norm differences

among ethnicities, difficulties with community agencies and possible ineffectiveness of

the program. This study adds to the existing literature of what works and what does not

work in dating violence prevention. Future dating violence prevention researchers should

take into account many of the hurdles encountered by this researcher and previous

researchers and make a plan in how to minimize the impact these hurdles have on the

results of the study. Further, more research is needed on how gender norm attitudes

change and effective ways to bring about this change.

Students also viewed