Athletic Coaches as Violence Prevention Advocates

Abstract

Adolescent relationship abuse (ARA) is a significant public health problem. Coaching Boys Into Men (CBIM) is an evidence-based ARA prevention program that trains coaches to deliver violence prevention messages to male athletes. Assessing acceptability and impact of CBIM on coaches may inform prevention efforts that involve these important adults in health promotion among youth. As part of a two-armed cluster-randomized controlled trial of CBIM in 16 high schools in Northern California, coaches completed baseline and postseason surveys (n = 176) to assess their attitudes and confidence delivering the program. Coaches in the intervention arm also participated in interviews (n = 36) that explored program acceptability, feasibility, and impact. Relative to controls, intervention coaches showed increases in confidence intervening when witnessing abusive behaviors among their athletes, greater bystander intervention, and greater frequency of violence-related discussions with athletes and other coaches. Coaches reported the program was easy to implement and valuable for their athletes. Findings illustrate the value of exploring attitudinal and behavioral changes among ARA prevention implementers, and suggest that coaches can gain confidence and enact behaviors to discourage ARA among male athletes. Coaches found the program to be feasible and valuable, which suggests potential for long-term uptake and sustainability.

Keywords

coaches high school male athletes dating violence sexual violence prevention gender-based violence bystander intervention

Adolescent relationship abuse (ARA) and sexual violence (SV) are major public health problems. Nearly one in three adolescent girls in the United States is a victim of physical, emotional, or verbal abuse by a dating partner (Davis, 2008). High school aged females are disproportionately affected by SV; more than one in eight high school girls report forced sex in their lifetime (Basile et al., 2006) and more than three quarters of women who have been sexually assaulted report their first such experience was before the age of 25 (Black et al., 2011; Masho, Odor, & Adera, 2005; Tjaden & Thoennes, 1998). Violence prevention programs that engage men and boys to adopt gender-equitable, nonviolent attitudes are increasingly recognized as a potential public health strategy to reduce violence against women and girls (Abbey & McAuslan, 2004; Barker, Ricardo, & Nascimento, 2007; Cleveland, Herrera, & Stuewig, 2003; Degue & DiLillo, 2004; Foshee, 1996; Foshee et al., 2004; Hines, 2007; International Center for Research on Women & Instituto Promundo, 2007; McMahon, 2010; Reed, 2008; Reed et al., 2008; Santana, Raj, Decker, La Marche, & Silverman, 2006; United Nations Population Fund (UNFPA) & Promundo, 2010; Warkentin & Gidycz, 2007).

Despite the prevalence of ARA/SV and increased support for gender-specific interventions, programs incorporating gender norms change and bystander intervention approaches (i.e., teaching youth to interrupt abusive behaviors among their peers) to prevent perpetration of ARA/SV are limited. Research-tested ARA/SV prevention programs generally require classroom instruction, competing with instruction of core curriculum and school personnel time. Nontraditional strategies for integration of violence prevention education into young people’s lives are needed. Nationally, more than 80% of school-age children participate in formal athletics programs, and participation continues to increase each year (The National Federation of State High School Association, 2012; Child Stats.Gov, 2012), making athletics an important out-of-classroom setting for prevention programming (Hilgers, 2006). Moreover, athletes are an important target for ARA/SV prevention, due to the prevalence of abuse perpetration among male athletes and their endorsement of attitudes supportive of violence against women (Boeringer, 1999; Moynihan, Banyard, Arnold, Eckstein, & Stapleton, 2011). Male athletes have also been noted to be regarded as leaders compared with nonathletes (Dobosz & Beaty, 1999), such that interventions with this population could influence the greater student body via athletes’ modeling of respectful behavior toward their dating partners (Fraser-Thomas, Cote, & Deakin, 2005; Jackson & Davis, 2000; Katz, Heisterkamp, & Fleming, 2011).

Prevention interventions with athletes necessitate involvement of coaches, a currently untapped resource, who have the opportunity to be consistent nonparental role models in the lives of students, interacting with athletes daily at practice and tournaments (Drewe, 2000; Martin, Richardson, Weiller, & Jackson, 2004). This high level of interaction provides an opportunity for coaches to develop close relationships with their athletes, foster trust as adult authority figures, and serve as resources for students (Cote & Salmela, 1996). Whether coaches are amenable to taking on this role in ARA/SV prevention and whether coaches’ attitudes and behaviors can be shifted toward greater involvement in violence prevention efforts with their athletes is not known. The current study aims to fill this gap by exploring coaches’ involvement in such violence prevention efforts with their athletes.

Coaching Boys Into Men (CBIM) is a research-tested ARA/SV prevention program intended to alter gender norms that foster ARA/SV perpetration, promote bystander intervention, and reduce ARA/SV perpetration (Miller et al., 2012; Miller et al., 2013). Coaches are guided to deliver messages to their athletes that focus on stopping violence against females via a series of 12 Training Cards, which provide 15-min scripted discussions to be delivered weekly throughout the athletic season (www.CoachesCorner.org). Lessons highlight respect, nonviolence, sexual consent, and interrupting abusive behaviors among peers.

Evaluation of this program with high school male athletes, which included three surveys over the course of a year as well as focus groups with a subsample of athletes, found short-term (end of sports season) improvements in intentions to intervene (intentions to stop abusive behaviors among peers) and positive bystander intervention behavior (proactive behavior in response to witnessing abusive behaviors among peers) as well as long-term (12 months post-baseline) reductions in negative bystander behavior and abuse perpetration against female dating partners (Miller et al., 2012; Miller et al., 2013). Evaluation of the impact of delivering the CBIM program on coaches is critical for several reasons. First, coaches were selected as the vehicle for program delivery because they are in a unique position to influence high school male athletes. As such, their delivery of the material is inevitably influenced by their own attitudes and experiences related to gender norms and ARA/SV. We hypothesized that delivering the program would improve coaches’ attitudes, confidence, and behaviors to prevent ARA/SV. Second, the success of the program hinges on coaches’ willingness to engage their athletes in ongoing conversations about violence in a meaningful way. Qualitative feedback from coaches about the acceptability and feasibility of conducting this program with their athletes can inform future implementation in other settings. Therefore, the purpose of the present study was to assess the impact of the CBIM program on high school athletic coaches’ attitudes and behaviors as well as to explore the acceptability and feasibility of working with coaches to discourage male athletes’ use of violence in their dating relationships and to encourage constructive intervention when witnessing such behaviors among their peers.

Method

Participants and Procedures

A total of 16 high schools in Northern California participated in a cluster-randomized controlled trial of CBIM (October 2009 to October 2011). Prior to randomization, each school’s athletic director approached head coaches of male and co-educational sports teams to encourage their participation in the study. At the start of each season (winter 2009-2010, spring and fall 2010), coaches were approached by study staff and 87% agreed to participate in the study. The primary participation barrier reported by coaches was lack of time.

After informed consent procedures, coaches completed a survey (Time 1) prior to the start of their sports season (n = 176). Coaches in intervention schools received the 60-min CBIM training and a Coaches Kit (containing training cards and resources for delivering the program). Coaches in the control schools continued with their usual standard coaching activities. Coaches then completed a follow-up survey immediately following the end of their sports season (Time 2, approximately 3 months after Time 1; n = 117, 66% retention). Online surveys were completed either on school computers simultaneously with athletes’ survey administration or a link was emailed to coaches to complete the survey on their own time. To facilitate anonymous matching of baseline and follow-up surveys, coaches self-created an identification code by responding to questions to which only they would know the answer.

Coaches did not receive personal compensation for their participation in the program delivery or surveys; instead, each school athletic department received a nominal stipend to thank them for allowing their sports teams to participate in the study.

At the end of each season (Time 2), coaches from the intervention arm were invited to participate in a semistructured face-to-face interview regarding their experiences with CBIM. Interested coaches completed additional informed consent procedures and a trained male interviewer conducted the interviews (n = 36) in a private space at the coach’s school. Coaches received a US$10 gift card to thank them for their time and effort participating in a face-to-face interview. Each interview was audio recorded and transcribed verbatim, checked for accuracy, any identifying information removed from transcripts, and audio files destroyed.

A mixed-methods design (i.e., combined quantitative and qualitative approaches) was chosen for the evaluation of the impact of CBIM on coaches because this allowed for assessment of changes in attitudes, confidence, and behaviors over time in a rigorous fashion with a comparison group of coaches, while interviews with coaches who implemented CBIM elucidated potential mechanisms for change (i.e., helping to contextualize the quantitative findings) as well as provided insights into the acceptability and feasibility of a coach-delivered violence prevention program for high school male athletes.

Study methods were approved by University of California Davis Human Subjects Committee, University of Pittsburgh Institutional Review Board, and by each school district.

Instruments

Quantitative surveys

Coaches’ demographic data included number of years coaching, age, race/ethnicity, gender, level of education completed, and whether they were born in the United States. Coaches reported which sports, gender, and age ranges of athletes they coached.

Three items asked about how often coaches had conversations with their athletes in the past 3 months about violence against women and girls, sexual harassment, and physical violence on and off the field. A fourth item asked about discussions they had with other coaches about the role of a coach in personal development of their athletes (Cronbach’s α = .77, items above).

Eleven questions were used to assess coaches’ gender attitudes, modified from the Gender-Equitable Norms Scale (Cronbach’s α = .78; Pulerwitz & Barker, 2008). For instance, items included the following: “Most boys do not stay faithful to their girlfriends for very long” and “If a girl is raped, it is often because she did not say no clearly enough.” A 5-point Likert-type scale ranged from strongly agree to strongly disagree. A mean score was calculated based on responses to the 11 items; a higher score indicated more gender-equitable attitudes.

Six items assessed coaches’ self-efficacy to have conversations with their athletes about violence against women (Cronbach’s α = .88). These questions included the following: “I know what I would say to a male athlete who is making sexual jokes that make fun of women and girls” or “I know how to talk to my team about preventing sexual assault.” Using a reverse-coded 5-point Likert-type scale of strongly agree to strongly disagree, a mean score was calculated with a higher score representing greater confidence handling such discussions with their athletes.

Coaches were asked to indicate whether they had witnessed any of nine abusive behaviors perpetrated by their athletes in the past 3 months. If they had witnessed a particular behavior, they were then asked how they responded to the behavior. Positive intervention included the following: (a) “I immediately reminded the athlete that such behavior is unacceptable,” (b) “I told the athlete in private that such behavior is unacceptable,” and (c) “I told the entire team (including this athlete) that such behaviors are unacceptable.” The response “I didn’t say anything” was coded as negative intervention. For each abusive behavior, separate binary indicators for “any positive” and for “any negative” bystander behavior were created. If an abusive behavior was not witnessed, both indicators were coded zero. These indicator variables were then summed to form positive and negative bystander intervention scores, which ranged from 0 (no intervention) to 9 (engaged in positive/negative bystander behavior in all 9 cases).

In the follow-up survey, intervention coaches were also asked about how they used the CBIM toolkit, and how often they discussed components of the toolkits with their athletes or with other coaches and school administrators.

Qualitative interviews

After delivery of the program and completion of follow-up surveys, intervention coaches were approached by research staff to participate in a semistructured face-to-face interview (n = 36). Of the coaches interviewed, most coached football, 31% (n = 11), and basketball, 14% (n = 5). Other sports coaches reported included wrestling (n = 3), baseball (n = 3), lacrosse (n = 1), soccer (n = 3), swimming (n = 1), track and field (n = 4), volleyball (n = 1), cross country (n = 1), and unknown sport (n = 3). Interview questions (supplemental appendix) explored how coaches used the CBIM toolkit, feedback about the program, observations from delivering the program, and thoughts about using CBIM in the future along with additional recommendations for other coaches using the program.

Data Analysis

Statistical methods

In light of the high intensity of within-school interactions among coaches and athletes, the unit of randomization for this study was the high school, and the unit of analysis was the coach. To account for the clustered randomized study design and the hierarchical arrangement of our data, a combination of survey data analysis methods and multilevel mixed-effects models in SAS/STAT were used (Hayes & Moulton, 2009; LaVange, Koch, & Schwartz, 2001; SAS Institute, 2009). For baseline comparisons on demographics, Fisher’s exact tests were used. Within-arm changes from baseline to follow-up in frequency of coaches witnessing abusive behavior were assessed with mixed-effects logistic regression models, stratified by study arm. For each of the other outcomes, between-arm differences were estimated in mixed-effects linear regression models to account for the cluster-randomized study design. Adjusted between-arm comparisons for mean differences from baseline to follow-up were assessed using mixed-effects models for longitudinal data that included covariates to adjust for age, race, gender, gender coached, years coaching, and education. These analyses included all available data from all participating coaches (176 baseline surveys, 117 follow-up surveys; DeSouza, Legedza, & Sankoh, 2009; Peters et al., 2012; Salim, Mackinnon, Christensen, & Griffiths, 2008). One coach participated in the program twice due to coaching multiple sports throughout the school year; a sensitivity analysis determined that including this coach did not substantively affect estimates or conclusions. In addition to survey data, a process evaluation was conducted, which included biweekly check in with coaches conducted by the school-site coordinator and coaches’ use of a tracking tool on which they recorded the cards they delivered and other program activities they completed with their athletes. Secondary analyses comparing outcomes based on the number of cards completed, including the intensity and length of time with intervention delivery, did not produce significantly different results from intention-to-treat analyses and are not presented here.

Two independent coders coded the interview transcripts using a directed content analysis approach, first with first with an a priori code list based on core questions with additional codes added as new content arose (Hsieh & Shannon, 2005). The principal investigator (PI) trained both coders on the coding process, providing the a priori list and guiding decisions about new codes emerging from the qualitative data. In collaboration with the PI, coded transcripts were then compared for agreement, the codebook finalized, and transcripts re-coded with the final codebook. Following this consensus coding process, codes related to implementation strategies were identified, and those chunks of text were analyzed for key themes. Additions of new codes or changes in code definitions were determined via consensus among the research team. The current analysis focuses on codes related to the role of coaches, their perceived role in violence prevention, their use of the CBIM program, and their perspectives on the impact of the program on themselves and their athletes. All qualitative analyses were conducted in the Altas.ti program (ATLAS.ti, 1997-2004).

Results

Demographic characteristics are presented in Table 1. Of the 176 coaches who completed baseline, 59 were lost to follow-up (33.5%), with more coaches lost to follow-up in intervention schools compared with controls. Those lost to follow-up were significantly (p < .05) more likely to coach male athletes only, be younger than 30, have coached less than a year, and have less than a college degree compared with coaches who completed the study. Those lost to follow-up were slightly less likely at baseline to have discussed youth development with other coaches. Comparing intervention and control coaches, there were no differences in baseline demographics or outcomes of interest (Table 2). Almost half of the coaches were from basketball or football teams as these are large teams with multiple assistant coaches involved in the CBIM program delivery.

Table 1.

Demographic Characteristics for the Total Sample, Intervention, and Control Arms.

	Total (n)	Intervention (n)	Control (n)
	n = 176	n = 99	n = 77
Age
<20 years old	2% (4)	4% (4)	0% (0)
20-29 years old	18% (32)	15% (15)	22% (17)
30-39 years old	31% (55)	32% (31)	31% (24)
40-49 years old	24% (42)	29% (28)	18% (14)
50-59 years old	18% (31)	15% (15)	20% (16)
>60 years old	6% (11)	5% (5)	8% (6)
Exact χ² p value			.2057
Race
White	51% (88)	47% (45)	57% (43)
Non-Hispanic Black	23% (39)	28% (27)	16% (12)
Hispanic	13% (22)	14% (13)	12% (9)
Asian	5% (8)	2% (2)	8% (6)
Pacific Islander	3% (5)	3% (3)	3% (2)
Native American	0% (0)	0% (0)	0% (0)
Multiracial	5% (8)	5% (5)	4% (3)
Other	0% (0)	0% (0)	0% (0)
Exact χ² p value			.2183
Gender
Female	8% (14)	8% (8)	8% (6)
Male	92% (158)	92% (89)	92% (69)
Exact χ² p value			.2203
Country of origin
Born in the United States	96% (165)	95% (92)	97% (73)
Born outside of the United States	4% (7)	5% (5)	3% (2)
Exact χ² p value			.2290
Education completed
Grade 9-11	<1% (1)	1% (1)	0% (0)
Grade 12 or GED (high school graduate)	4% (7)	6% (6)	1% (1)
Some college/technical school	25% (44)	26% (25)	25% (19)
Graduated from college or technical school	32% (56)	33% (32)	31 % (24)
Completed graduate school	38% (65)	33% (32)	43% (33)
Exact χ² p value			.3345

Note. Do not all add up to 100% due to small amounts of missing data (<4%).

Table 2.

Coaching Characteristics for the Total Sample, Intervention, and Control Arms.

	Total (n)	Intervention (n)	Control (n)
	n = 176	n = 99	n = 77
Number of years coaching
<1 year	6% (11)	5% (5)	8% (6)
1-5 years	25% (44)	30% (29)	19% (15)
5-10 years	22% (40)	20% (20)	26% (20)
>10 years	46% (80)	45% (44)	47% (36)
Exact χ² p value			.4092
Gender coached
Male only	40% (70)	44% (42)	36% (28)
Female only	1% (2)	2% (2)	0% (0)
Both males and females	58% (101)	54% (52)	64% (49)
Exact χ² p value			.2692
Ages coached
College graduates	3% (6)	3% (3)	4% (3)
College students	3% (5)	3% (3)	3% (2)
High school students	98% (172)	99% (98)	96% (74)
Middle school students	15% (27)	17% (17)	13% (10)
Elementary school students	9% (15)	9% (9)	8% (6)
Sport coached
Basketball	15% (27)	12% (12)	19% (15)
Wrestling	9% (16)	10% (10)	8% (6)
Baseball	9% (15)	8% (8)	9% (7)
Golf	2% (3)	1% (1)	3% (2)
Lacrosse	1% (1)	1% (1)	0% (0)
Rugby	1% (2)	0% (0)	3% (2)
Soccer	9% (16)	9% (9)	9% (7)
Swimming	2% (4)	2% (2)	3% (2)
Tennis	5% (8)	5% (5)	4% (3)
Track and field	2% (4)	4% (4)	0% (0)
Volleyball	5% (8)	3% (3)	6% (5)
Water polo	2% (3)	0% (0)	4% (3)
Cross country	6% (11)	5% (5)	8% (6)
Football	31% (55)	38% (38)	22% (17)

Note. Do not all add up to 100% due to small amounts of missing data (<4%).

Intervention coaches reported witnessing more abusive behaviors among their athletes compared with control coaches at follow-up (Table 3). In adjusted analyses, intervention coaches showed a significantly increased mean change from baseline to follow-up in positive bystander intervention behavior (0.82, 95% CI [0.16, 1.48]; Table 4) and confidence to address athletes about abusive behaviors toward women and girls (0.70, 95% CI [0.48, 0.91]; Table 4) compared with control coaches. Coaches in the intervention arm reported a greater number of discussions with their athletes about SV against women and girls and the use of physical violence on and off the field (0.89, 95% CI [0.60, 1.19]; Table 4) as well as the number of discussions with other coaches about the role of a coach in the personal development of their athletes (0.36, 95% CI [0.002, 0.72]; Table 4). Post hoc analyses that included adjustment for the intensity of intervention delivery (i.e., how often coaches discussed the training cards with their athletes) did not substantially alter these findings (results not shown).

Table 3.

Coaches Witnessing Athletes’ Abusive Behaviors, % (N).

Witnessed Abusive Behaviors	Intervention (All)		Control
Witnessed Abusive Behaviors	Baseline (n = 99)	Follow-Up (n = 64)	Baseline (n = 77)	Follow-Up (n = 53)
A male athlete making rude or disrespectful comments about a girl’s body, clothing, or make-up, such as catcalling or jeering	47.4 (46)	61.9 (39)	38.2 (29)	34.6 (18)
A male athlete spreading rumors about a girl’s sexual reputation, like saying she’s “easy”	20.4 (20)	23.4 (15)	11.7 (9)	11.5 (6)
A male athlete telling a girl who she can talk to or hang out with	7.1 (7)	10.9 (7)	6.5 (5)	1.9 (1)
A male athlete telling sexual jokes that disrespect women and girls	28.9 (28)	44.4 (28)^†	28.9 (22)	21.6 (11)
A male athlete bragging about what he got a girl to do sexually	16.7 (16)	25 (16)^†	10.5 (8)	5.8 (3)
A male athlete showing other people sexual messages about a girl, or nude/sexual pictures of a girl on a cell phone or the Internet	9.2 (9)	12.7 (8)	2.6 (2)	5.9 (3)
A male athlete doing unwelcome or uninvited things toward a girl (or a group of girls) such as howling, whistling or making sexual gestures	24.5 (24)	40.6 (26)*	20.8 (16)	25 (13)
A male athlete fighting with a girl where he’s starting to cuss at or threaten her	14.3 (14)	9.4 (6)	9.1 (7)	7.7 (4)
A male athlete shoving, grabbing, or otherwise physically hurting a girl	6.1 (6)	7.8 (5)	5.2 (4)	7.8 (4)
Witnessing any of the above abusive behaviors	60.2 (59)	76.6 (49)⁺	55.8 (43)	51.9 (27)

Note. N listed in header is number of surveys in each group; missing values for individual questions bring the reported ns down by one or two per group. For each behavior, separate mixed-effects logistic regression models were fit for each study arm to test within-arm baseline to follow-up differences in proportion of coaches witnessing the behavior, using random effects for coaches to account for clustered study design.

†

p < .10 for within-arm baseline to follow-up differences. *p < .05 for within-arm baseline to follow-up differences.

Table 4.

Baseline and Follow-Up Means and Standard Deviations for Outcomes of Interest Among Intervention/Control Coaches and Regression Adjusted Intervention Effects on Mean Improvements From Baseline to Follow-Up.

	Baseline			Follow-Up		Adjusted Analyses
	Intervention	Control	Intervention vs. Control	Intervention	Control	Adjusted Intervention
	M (SD)	M (SD)	p	M (SD)	M (SD)	Effect (95% CI)
Positive intervention score	1.40 (1.83)	1.18 (1.72)	.55	2.27 (2.12)	1.11 (1.67)	0.82 [0.16, 1.48] *
Negative intervention score	0.32 (0.93)	0.14 (0.39)	.16	0.08 (0.32)	0.08 (0.26)	−0.18 [−0.37, 0.02]
Gender attitude (1-5)	3.20 (0.55)	3.33 (0.56)	.20	3.37 (0.52)	3.50 (0.63)	0.02 [−0.14, 0.17]
Coach confidence (1-5)	3.88 (0.64)	4.04 (0.60)	.27	4.50 (0.47)	4.00 (0.61)	0.70 [0.48, 0.91] ***
Discussion with athletes (1-4)	2.15 (0.82)	2.03 (0.78)	.53	3.48 (0.5)	2.36 (0.76)	0.89 [0.60, 1.19] ***
Discussion with coaches (1-4)	2.80 (1.09)	2.78 (1.06)	.94	3.33 (0.61)	2.89 (0.86)	0.36 [0.002, 0.72] *

Note. p values for between-arm differences in baseline means were estimated in simple mixed-effects linear regression models of baseline data, with random effects for schools, to account for the cluster-randomized study design. Adjusted intervention effects and 95% confidence intervals (CI) describe the adjusted mean between-arm difference in baseline-adjusted follow-up scores and were estimated in mixed-effects multiple linear regression models of baseline and follow-up data, with covariates to statistically adjust for age group, race, gender coached, years coaching, gender, education, and previous training in any one of the following: sexual harassment, dating violence, or sexual assault. These mixed-effects models included random effects for schools and for coaches nested within schools to account for study design.

p<0.05

p<0.01

***

p<0.001

For each behavior, separate mixed-effects logistic regression models were fit for each study arm to test within-arm baseline to follow-up differences in proportion of coaches witnessing the behavior, using random effects for coaches to account for clustered study design.

Interviews

Implementation

Coaches interviewed were overwhelmingly positive about the ease of delivering the CBIM training cards, indicating they were “helpful and easy to use,” they “[didn’t] take tons of time,” and “it was really cool to have a structured thing.” The most commonly mentioned card in the series addressed bragging about one’s sexual reputation. Coaches reported the most lively discussions with this card:

[Athletes] understood about relationship abuse . . . they knew it’s not just physical, it’s also mental and emotional abuse . . . but what they didn’t really have a clear picture of . . . what they felt maybe was harmless: bragging about their sexual self . . . making dirty jokes . . . they hadn’t made a connection that could be harmful.

Role of coaches in violence prevention

Some coaches reported being reticent to do the program due to time limitations:

When you first [talked] to me about this I didn’t want to do it because time is super precious . . . but I talked to [another coach] who did [CBIM] he said it was good, it was really good, it was worth doing. He said if I could work it into the football schedule we should do it.

Coaches noted changes in their attitudes with implementation, recognizing that educating their athletes about respectful behavior was one of their roles: “This had to be talked about before the issue occurs” and “It’s so easy to put [CBIM] off. We’re focused on our game. It’s not until you start [to] implement it you realize it’s not a task but an opportunity.” These coaches also appreciated that discussing these topics was critical because students did not get the material in other ways. One coach stated, “The topics are necessary . . . We don’t have a class that teaches this.” Another said, “Our kids need this kind of program. A lot of them don’t have very good role models and good examples . . . of how you are supposed to treat a girl.”

One coach also disclosed that this program was deeply personal and informed his commitment to the program: “I used to see my mom get knocked around some and then once I got big enough to stand between him, I got knocked around.”

Finally, coaches recognized the lasting impact they may have on their athletes and the ability of CBIM to contribute to that impact:

Sometimes we worry about wins and losses, but if you’re doing all the right things, wins and losses take care of themselves. The bottom line is we’re here to make our student athletes better citizens and . . . get them to the next level whether they play a sport or go to college, or join the workforce. The most ultimate thing that I find as a coach is when you see these young people later on in life, they’re being successful and you realize they’re being successful because of something they’ve gotten through your program, whatever it is, and in athletics we have this ability to affect them in many ways and the Coaching Boys to Men program, if we look back 5, 10 years from now, we’re gonna find out—that made a big difference in a lot of our young men’s lives.

Impact of the program on athletes from coaches’ perspectives

Coaches uniformly reported changes in athletes’ language, attitudes, and behaviors after participating in CBIM. One coach described his athletes before the program:

There was a girl walking by the practice facility and the guys were looking at her disrespectfully, and I could tell the girl took a left-hand turn early before she went into the direction she really wanted to go. It just brought the program to life right there about disrespect . . . the seriousness of the effects of [athletes’] actions.

After the program, athletes used more respectful language toward their peers:

The teasing changed . . . The language used . . . calling each other girls . . . there were always a couple of boys in here that used to always say ‘come on ladies’ and now they are correcting [each other].

Some coaches highlighted the challenges boys face in intervening when they witness peers engaging in disrespectful or harmful behaviors, and the challenges teaching youth these skills.

The one thing that all the boys were really hesitant about is “what if you see one of your teammates interacting inappropriately with their girlfriend? What would you do?” . . . I encouraged them that you really do need to step up and say something. They were all like “but coach, we’re going [to get] fired on if we do.” . . . So that is a concern of the boys. Especially in today’s society, to get involved in that kind of situation really does put them in harm’s way.

Despite these challenges, coaches noted an overall change in the team’s behaviors as a whole: “It brought us closer as a unit. I had young men that held each other accountable.” CBIM became a program that was an important part of their team experience:

I think it’s a great program. There were times during track practice during warm-ups, when I would make a remark jokingly, and one of the guys would go, “Hey, That’s not very good Boys to Men thing here is it?” Even though they knew I was joking about it, it was their way of saying, “hey, I remembered that.”

Discussion

This mixed-methods study examined the impact of the CBIM program on high school athletic coaches’ attitudes and behaviors. Findings suggest coaches found the program valuable for their athletes and the card series easy to use for delivery of violence prevention messages. Intervention coaches demonstrated significant increases in positive bystander intervention, confidence intervening with athletes, frequency of violence prevention discussions with athletes, and frequency of program discussion with other coaches compared with controls. This brief curriculum appears to impact both coach confidence to enact behavior change and actual positive behavior changes. Whether these attitude and behavior changes among coaches can be sustained longer term remains to be seen. Nonetheless, the enthusiasm among coaches for the program and their willingness to discuss stopping violence against women with their athletes identified in this study suggest that the program’s impact on coaches could translate into continued implementation of the program into subsequent coaching seasons.

The CBIM program is intended to alter norms that foster ARA/SV perpetration by engaging athletic coaches as positive role models to deliver violence prevention messages to adolescent male athletes. Findings affirm the feasibility of engaging coaches as nonparental role models to positively influence male athletes’ use of violence in their dating relationships. The observed changes in coach behavior likely underpin the positive impact of CBIM among athletes (Miller et al., 2012; Miller et al., 2013). With delivery of CBIM, coaches had a heighted awareness (evidenced in the increased likelihood of witnessing athletes’ abusive behaviors) and were better equipped to identify and prevent abusive behaviors among their athletes. Literature suggests that coaches are in a position to alter athletes’ attitudes and behaviors (Poczwardowski, Barott, & Henschen, 2002), with the dynamic coach–athlete relationship that models both sports-specific skills and social development (Poczwardowski, Barott, & Henschen, 2002; Poczwardowski, Barott, & Peregoy, 2002). Coaches are invested in the personal development of their athletes (Cote & Salmela, 1996), teaching values such as hard work and life skills (McCallister, Blinde, & Weiss, 2000). In our sample, coaches articulated the role they play in the personal development of their athletes in addition to success on the field, which renders them ideal implementers of CBIM.

The control arm coaches also reported having some discussion with their athletes about respect and nonviolence. The comparative increase in these discussions between control and intervention coaches could be attributed to intervention coaches receiving the CBIM training and delivering this program, that is, demonstrating intervention implementation as intended. While these differences in the frequency of conversations with athletes about nonviolence seem intuitive and expected, they represent one potential mechanism through which the CBIM program likely influences athletes and are important findings given that the program has been shown to reduce abuse perpetration among high school athletes.

A majority of coaches interviewed, potentially a nonrepresentative sample, found this violence prevention program to be valuable and easy to implement with their athletes. Coaches acknowledged they may have previously underestimated the influence they can have on their athletes and how doing the program contributed to a visible positive impact. Coaches identified changes in athletes’ language, attitudes, and behaviors, which may also influence peers and the school climate in the long term. However, coaches also identified challenges their athletes face intervening with their peers’ abusive behaviors including fear of retribution from classmates for intervening. Next steps could include additional training for coaches on how to address such challenges with their athletes to increase positive bystander intervention among youth.

Limitations

These findings should be interpreted in light of several limitations. First, reliance on self-report of self-efficacy, actual frequency of discussions with coaches and athletes, and bystander intervention may result in overestimates of these behaviors due to social desirability biases. Although the coaches created a self-generated anonymous code for computerized survey administration, which was intended to reduce misreporting of sensitive items, given the small sample size, the anonymous code may not have eliminated social desirability bias. While it is possible that coaches who were trained in CBIM would be more inclined to report intervening positively because they are aware of the expectation to intervene as part of implementing this program, this differential reporting (i.e., intervention coaches being inclined to give the right answer) could be interpreted as support for the intervention, given that the program operates through shifting social norms. Second, the sample size of coaches who completed surveys and interviews was small, and likely biased toward coaches who felt positively about the program. Finally, whether intervention effects last beyond the immediate postseason remains to be seen.

Conclusion

These limitations notwithstanding, the acceptability, feasibility, and value of the CBIM program as indicated by participating coaches points to the potential value of engaging coaches as allies in violence prevention and health promotion efforts in school and community settings. Their role as influential, nonparental role models provides a unique avenue for coaches to positively impact how young male athletes think and behave.

Footnotes

Acknowledgements

We wish to thank the school districts, high school principals, athletic directors, coaches, school administrators, staff, students, and the advocates at WEAVE for their enthusiastic participation in this study.

Authors’ Note

The study sponsor did not have any influence on study design, data collection, analysis, interpretation, manuscript preparation, nor decision to submit for publication. The views expressed in this manuscript are solely the responsibility of the authors, and do not represent the views of the Centers for Disease Control and Prevention. Preliminary findings from this study have been presented at the American Public Health Association meeting in 2010 and 2011, Healthy Teen Network in 2011, and the Society for Adolescent Health and Medicine in 2011 and 2012. Clinical Trials Registration: “Evaluation of ‘Coaching Boys Into Men’ (CBIM) Program” NCT01367704 (at ).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Centers for Disease Control and Prevention (Grant CE001561-01).

Author Biographies

Maria Catrina D. Jaime completed her master’s in public health at University of California Davis. Currently, she is the Senior Clinical Research Coordinator of Adolescent Medicine at Children’s Hospital of Pittsburgh, the University of Pittsburgh Medical Center. She has been the project manager of the Coaching Boys Into Men (CBIM) research study and worked closely with athletes and coaches in 16 public high schools in Sacramento, California. She is involved in development of CBIM for middle schools and the CBIM national program evaluation. As a liaison to the Community PARTners Core of the Clinical Translational Science Institute at the University of Pittsburgh, she also works to increase involvement of community members in community–academic partnerships. She is currently a doctoral student at the department of Community and Behavioral Health Sciences at the Graduate School of Public Health, University of Pittsburgh.

Heather L. McCauley is an assistant professor of Pediatrics in the Division of Adolescent Medicine at Children’s Hospital of Pittsburgh of University of Pittsburgh Medical Center. She holds a ScD in social epidemiology and a MS in global health from Harvard University. Her research focuses on the health impacts of adolescent relationship abuse and sexual assault. Her current work looks at the intersection of adverse childhood experiences, adolescent substance use, and sexual assault in vulnerable populations. She was also the data analyst for the school-based violence prevention intervention CBIM a programmatic example of how she works to incorporate gender analysis in her research.

Daniel J. Tancredi is a biostatistician and assistant professor in the Department of Pediatrics and a member of the Center for Healthcare Policy and Research at the University of California, Davis. He has expertise in statistical methods for the design and analysis of experimental trials of behavioral interventions and observational studies and more than 20 years of experience heavily involved in collaborative medical and survey research at University of California, Davis, and at Rush University in Chicago. He holds a bachelor’s degree in behavioral sciences from the University of Chicago and a PhD in mathematical statistics from the University of Illinois at Chicago.

Jasmine Nettiksimmons is a postdoctoral fellow in the Department of Psychiatry at University of California, San Francisco. She holds a PhD in epidemiology from the University of California, Davis, and a MA in mathematics/statistics from the University of Montana. She has more than 5 years of experience performing statistical analyses for clinical and behavioral health-related research.

Michele R. Decker, ScD, MPH, is an assistant professor in the Department of Population, Family and Reproductive Health at the Johns Hopkins Bloomberg School of Public Health. Trained as a social epidemiologist, her research addresses gender-based violence through both primary prevention and intervention, as well as its sexual and reproductive health implications. Much of this work is focused on marginalized populations, including urban men and women, adolescents, and those involved in transactional sex or sex work.

Jay G. Silverman, PhD, is co-director of the Center on Gender Inequities and Global Health and a professor of Medicine and Global Public Health at University of California, San Diego. He is a leading global researcher on understanding and preventing gender-based violence against adolescent and adult women (e.g., intimate partner violence, sexual violence and sex trafficking). He has led multiple major federally, foundation and UN-funded studies of gender-based violence and its effects on the health of women and their children. He has led groundbreaking studies of sex trafficking in South and Southeast Asia, and has recently expanded this line of research to Latin America. His ongoing work in South Asia focuses on the implications of husband and in-law violence and nonviolent maltreatment for maternal and child health. He also continues to work domestically, developing and evaluating interventions to address chronic threats to adolescent girls’ sexual safety as well as reproductive coercion and intimate partner violence among adolescent and adult women.

Brian O’Connor holds a master’s in journalism from Columbia University and is a member of the National Association of Black Journalists. He is the director of Public Education Campaigns and Programs for Futures Without Violence. In this role, he crafts national and international violence prevention public awareness and action campaigns for the organization, most of which focus on reaching men, parents, coaches, teachers, military families, and teens. To date, his efforts have enabled local adaptations of these programs in countless communities around the world. In particular, his work engaging men and boys to help end violence against women was instrumental in the development of the international CBIM initiative in partnership with UNICEF in 2006 and continues today.

Nicholas Stetkevich is a medical student at Touro University California, College of Osteopathic Medicine. He has served as a clinic director at the school’s Student Run Free Clinic managing day-to-day operations. In addition, he has been a basic sciences tutor for the college of medicine and is a member of Sigma Sigma Phi, a community service-centered osteopathic honor society. Before medical school, he worked as a high school science teacher in San Jose, California, and in medical research on the CBIM study at University of California Davis School of Medicine. He graduated with a BS in biology from Santa Clara University, earned teaching credentials through San Jose State University, and completed his MPH at University of California, Davis.

Elizabeth Miller is chief of adolescent medicine and associate professor of pediatrics at Children’s Hospital of Pittsburgh, the University of Pittsburgh Medical Center. Trained in medical anthropology as well as internal medicine and pediatrics, her research has included examination of sex trafficking among adolescents in Asia, teen dating abuse, and reproductive health, with a focus on underserved youth populations including pregnant and parenting teens, foster, homeless, and gang-affiliated youth. Her research focuses on the impact of gender-based violence on young women’s health. She conducts research on brief clinical interventions to reduce partner violence and unintended pregnancy, funded by the National Institute of Child Health and Human Development and the National Institute of Justice. In addition, she is conducting a study of a sexual violence prevention program titled “Coaching Boys Into Men,” which involves training coaches to talk to their young male athletes about stopping violence against women, funded by the Centers for Disease Control and Prevention. She is involved in projects to reduce gender-based violence and improve adolescent and young adult women’s health in India and Japan.

References

Abbey

McAuslan

(2004). Longitudinal examination of male college students perpetration of sexual assault. Journal of Consulting and Clinical Psychology, 72, 747-756.

ATLAS.ti. (1997-2004). Atlas.ti: The Knowledge Workbench (Version 5.0). Berlin, Germany: Thomas Muhr, Scientific Software Development.

Barker

Ricardo

Nascimento

(2007). Engaging men and boys in changing gender-based inequity in health: Evidence from programme interventions. Geneva, Switzerland: World Health Organization.

Basile

K. C.

Black

M. C.

Simon

T. R.

Arias

Brener

N. D.

Saltzman

L. E.

(2006). The association between self-reported lifetime history of forced sexual intercourse and recent health-risk behaviors: Findings from the 2003 National Youth Risk Behavior Survey. Journal of Adolescent Health, 39, e751-e757.

Black

M. C.

Basile

K. C.

Breiding

M. J.

Smith

S. G.

Walters

M. L.

Merrick

M. T.

. . .Stevens

M. R.

(2011). The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 Summary Report. Atlanta, GA: National Center for Injury Prevention and Control Centers for Disease Control and Prevention.

Boeringer

S. B.

(1999). Associations of rape-supportive attitudes with fraternal and athletic participation. Violence Against Women, 5, 81-89.

Child Stats.Gov Forum on Child and Family Statistics. (2012). POP 1 Child population: Number of children (in millions) ages 0-17 in the United States by age, 1950-2012 and projected 2013-2050. Retrieved from http://www.childstats.gov/americaschildren/tables/pop1.asp

Cleveland

H. H.

Herrera

V. M.

Stuewig

(2003). Abusive males and abused females in adolescent relationships: Risk factor similarity and dissimilarity and the role of relationship seriousness. Journal of Family Violence, 18, 325-339.

Cote

Salmela

J. H.

(1996). The organizational tasks of high-performance gymnastic coaches. Sport Psychologist, 10, 247-260.

10.

Davis

(2008). Interpersonal and physical dating violence among teens. Oakland, CA: The National Council on Crime and Delinquency.

11.

Degue

DiLillo

(2004). Understanding perpetrators of nonphysical sexual coercion: Characteristics of those who cross the line. Violence and Victims, 19, 673-688.

12.

DeSouza

C. M.

Legedza

A. T.

Sankoh

A. J.

(2009). An overview of practical approaches for handling missing data in clinical trials. Journal of Biopharmaceutical Statistics, 19, 1055-1073.

13.

Dobosz

R. P.

Beaty

L. A.

(1999). The relationship between athletic participation and high school student’s leadership ability. Adolescence, 34, 215-220.

14.

Drewe

S. B.

(2000). Coaches, ethics and autonomy. Sport Education and Society, 5, 147-162.

15.

Foshee

V. A.

(1996). Gender differences in adolescent dating abuse prevalence, types and injuries. Health Education Research, 11, 275-286.

16.

Foshee

V. A.

Bauman

K. E.

Ennett

S. T.

Linder

G. F.

Benefield

Suchindran

(2004). Assessing the long-term effects of the Safe Dates program and a booster in preventing and reducing adolescent dating violence victimization and perpetration. American Journal of Public Health, 94, 619-624.

17.

Fraser-Thomas

J. L.

Cote

Deakin

(2005). Youth sport programs: An avenue to foster positive youth development. Physical Education & Sport Pedagogy, 10, 19-40.

18.

Hayes

Moulton

(2009). Cluster randomised trials. Boca Raton, FL: Chapman & Hall.

19.

Hilgers

(2006, July 5). Youth sports drawing more than ever. CNN. Retrieved from http://articles.cnn.com/2006-07-03/us/rise.kids.sports_1_youth-sports-tennis-lessons-kids?_s=PM:US

20.

Hines

D. A.

(2007). Predictors of sexual coercion against women and men: A multilevel, multinational study of university students. Archives of Sexual Behavior, 36, 403-422.

21.

Hsieh

H. F.

Shannon

S. E.

(2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15, 1277-1288.

22.

International Center for Research on Women & Instituto Promundo. (2007). Engaging men and boys to achieve gender equality: How can we build on what we have learned? Washington, DC: International Center for Research on Women.

23.

Jackson

T. L.

Davis

J. L.

(2000). Prevention of sexual and physical assault toward women: A program for male athletes. Journal of Community Psychology, 28, 589-605.

24.

Katz

Heisterkamp

H. A.

Fleming

W. M.

(2011). The social justice roots of the mentors in violence prevention model and its application in a high school setting. Violence Against Women, 17, 684-702.

25.

LaVange

L. M.

Koch

G. G.

Schwartz

T. A.

(2001). Applying sample survey methods to clinical trials data. Statistics in Medicine, 20, 2609-2623.

26.

Martin

S. B.

Richardson

P. A.

Weiller

K. H.

Jackson

A. W.

(2004). Role models, perceived sport encouragement, and sport expectancies of United States adolescent athletes and their parents. Women in Sport & Physical Activity Journal, 13, 18-27.

27.

Masho

S. W.

Odor

R. K.

Adera

(2005). Sexual assault in Virginia: A population-based study. Womens Health Issues, 15, 157-166.

28.

McCallister

S. G.

Blinde

E. M.

Weiss

W. M.

(2000). Teaching values and implementing philosophies: Dilemmas of the youth sport coach. Physical Educator, 57, 35-45.

29.

McMahon

(2010). Rape myth beliefs and bystander attitudes among incoming college students. Journal of American College Health, 59, 3-11.

30.

Miller

Tancredi

D. J.

McCauley

H. L.

Decker

M. R.

Virata

M. C.

Anderson

H. A.

. . .Silverman

J. G.

(2012). “Coaching Boys Into Men”: A cluster-randomized controlled trial of a dating violence prevention program. Journal of Adolescent Health, 51, 431-438.

31.

Miller

Tancredi

D. J.

McCauley

H. L.

Decker

M. R.

Virata

M. C.

Anderson

H. A.

. . .Silverman

J. G.

(2013). One-year follow-up of a coach-delivered dating violence prevention program: A cluster randomized controlled trial. American Journal of Preventive Medicine, 45, 108-112.

32.

Moynihan

M. M.

Banyard

V. L.

Arnold

J. S.

Eckstein

R. P.

Stapleton

J. G.

(2011). Engaging intercollegiate athletes in preventing and intervening in sexual and intimate partner violence. Journal of American College Health, 59, 197-204.

33.

Peters

S. A.

Bots

M. L.

den Ruijter

H. M.

Palmer

M. K.

Grobbee

D. E.

Crouse

J. R.

III . . .Koffijberg

(2012). Multiple imputation of missing repeated outcome measurements did not add to linear mixed-effects models. Journal of Clinical Epidemiology, 65, 686-695.

34.

Poczwardowski

Barott

J. E.

Henschen

K. P.

(2002). The athlete and coach: Their relationship and its meaning. Results of an interpretive study. International Journal of Sport Psychology, 33, 116-140.

35.

Poczwardowski

Barott

J. E.

Peregoy

J. J.

(2002). The athlete and coach: Their relationship and its meaning. Methodological concerns and research process. International Journal of Sport Psychology, 33, 98-115.

36.

Pulerwitz

Barker

(2008). Measuring attitudes toward gender norms among young men in Brazil: Development and psychometric evaluation of the GEM Scale. Men and Masculinities, 10, 322-338.

37.

Reed

(2008). Intimate partner violence: A gender-based issue? American Journal of Public Health, 98, 197-198.

38.

Reed

Silverman

J. G.

Raj

Rothman

E. F.

Decker

M. R.

Gottlieb

. . . Miller

(2008). Social and environmental contexts of adolescent and young adult perpetrators of intimate partner violence. American Journal of Men’s Health, 2, 260-272.

39.

Salim

Mackinnon

Christensen

Griffiths

(2008). Comparison of data analysis strategies for intent-to-treat analysis in pre-test-post-test designs with substantial dropout rates. Psychiatry Research, 160, 335-345.

40.

Santana

M. C.

Raj

Decker

M. R.

La Marche

Silverman

J. G.

(2006). Masculine gender roles associated with increased sexual risk and intimate partner violence perpetration among young adult men. Journal of Urban Health, 83, 575-585.

41.

SAS Institute. (2009). SAS (Version SAS v. 9.2). Cary, NC: Author.

42.

The National Federation of State High School Association. (2012). 2011-12 High School Athletics Participation Survey. Retrieved from http://www.nfhs.org/content.aspx?id=3282

43.

Tjaden

Thoennes

(1998). Prevalence, incidence and consequences of violence against women: Findings from the national violence against women survey (Department of Justice, National Institute of Justice, Ed.). Washington, DC: U.S. Department of Justice, National Institute of Justice.

44.

United Nations Population Fund (UNFPA) & Promundo. (2010). Engaging men and boys in gender equality and health. New York, NY UNFPA.

45.

Warkentin

J. B.

Gidycz

C. A.

(2007). The use and acceptance of sexually aggressive tactics in college men. Journal of Interpersonal Violence, 22, 829-850.