Coach Expectations About Off-Field Conduct and Bystander Intervention by U.S. College Football Players to Prevent Inappropriate Sexual Behavior

Abstract

The objective of the present study was to assess whether there is a positive association between expectations about off-field conduct set by the team coach and the likelihood that college football players intend to engage as prosocial bystanders in the prevention of what they consider to be inappropriate sexual behavior. In a sample of U.S. collegiate football players (N = 3,281), a path analysis model tested the association between coach expectations, perceived likelihood of discipline for off-field transgressions, and likelihood of intending to intervene to prevent inappropriate sexual behavior. Mediation of these relationships by the athlete’s sense of exploitative entitlement and their attitudes about intervening were also assessed. Findings supported the hypothesized relationships, with expectations and discipline associated with bystander intentions both directly and indirectly through the mediating pathways of entitlement and attitudes about intervening. These findings provide evidence about the important role that sports team coaches can play in encouraging bystander intervention by clarifying expectations and consequences for conduct off the field of play. Athletic departments can provide a framework within which coaches are informed about the importance of setting and enforcing standards for off-field behavior, and are appropriately incentivized to do so.

Keywords

prevention sexual assault intervention cultural contexts sport

Sexual violence is increasingly considered a critical problem on many American college campuses. Among college women, recent estimates of lifetime incidence of unwanted sexual contact, including rape, attempted rape, sexual assault, and sexual coercion range from 15% to 73% (Cleere & Lynn, 2013; McCauley, Ruggiero, Resnick, Conoscenti, & Kilpatrick, 2009; Mouilso & Fischer, 2012). Academic year estimates for incidence of unwanted sexual contact range from 12% to 20% (Banyard, Ward, et al., 2007; Messman-Moore et al., 2008). Characteristics of the social environment at many U.S. colleges, such as high levels of alcohol consumption (Cole, 2006; Palmer, McMahon, Rounsaville, & Ball, 2010) and norms about casual sex outside of committed relationships (England & Thomas, 2006), may increase the risk of unwanted sexual contact (Koelsch, Brown, & Boisen, 2012). Consequently, colleges are a critical venue for prevention efforts.

Consistent with multilevel models of behavior such as Bronfenbrenner’s (1979) Social Ecological Theory, a community approach to sexual violence prevention that engages peer bystanders in prosocial behavior has been recommended (Banyard, 2011; Banyard, Cross, & Modecki, 2006; Banyard, Plante, & Moynihan, 2004; Burn, 2009; McMahon & Banyard, 2012). Latane and Darley (1970) first outlined a series of stages through which bystanders pass when deciding whether or not to intervene: noticing a given situation, classifying it as a problem, accepting responsibility for action, selecting a course of action, and then engaging in that action. Part of this rational and deliberative process is a calculation of costs and benefits of intervening, with those perceiving greater costs relative to benefits less likely to intervene (Banyard, 2011; Banyard & Moynihan, 2011; Bénabou & Tirole, 2004).

These stages have informed the development of bystander-focused sexual violence prevention programming (Banyard, Moynihan, & Crossman, 2009; Banyard et al., 2007; Chaurand & Brauer, 2008; Gidycz, Orchowski, & Berkowitz, 2011; Miller et al., 2013; Miller et al., 2012; Moynihan & Banyard, 2008, 2011; Moynihan, Banyard, Arnold, Eckstein, & Stapleton, 2010; Vladutiu, Martin, & Macy, 2011). Some of this programming has targeted specific groups, based on evidence—or public perception—of that group’s elevated risk of perpetrating sexual violence. Although data are equivocal about whether members of high-profile male sports teams perpetuate sexual violence at a rate that is greater than their nonathlete campus peers (Boeringer, 1999; Chandler, Dewayne, & Carroll, 1999; Murnen & Kohlman, 2007; Sawyer, Thompson, & Chicorelli, 2002), context-specific prevention efforts are warranted among all groups of college students who are at risk of perpetrating sexual violence; consequently, understanding determinants of behavior related to the prevention of sexual violence among male collegiate athletes is an important area for study, but certainly not the only population among whom this work is necessary. In addition, as often highly visible members of the campus community, male athletes in high-profile sports who function as prosocial bystanders in the prevention of sexual violence have the potential to be influential opinion leaders (Banyard et al., 2009).

To develop effective programming to encourage bystander intervention in the prevention of sexual violence, it is critical that we understand what modifiable factors make members of the population of interest more or less likely to intervene. Despite numerous suggestions about the importance of multilevel and ecological approaches to prevention (e.g., Armstrong, Hamilton, & Sweeney, 2006; Casey & Lindhorst, 2009; McMahon & Banyard, 2012), most research on variability in bystander behavior has focused on individual attitudes and perceived peer attitudes (e.g., Banyard & Moynihan, 2011; Brown & Messman-Moore, 2010). Little research to date has explored the extent to which modifiable contextual factors are associated with the individual’s attitudes and intentions about engaging in bystander behavior.

According to the Theory of Planned Behavior (TPB; Ajzen, 1991), a direct predictor of whether or not individuals perform a given behavior is whether they intend to perform that behavior. Although this relationship is supported empirically, a substantial proportion of behavior is often unexplained by intentions (Webb & Sheeran, 2006), with contextual influences and an individual’s emotional and physiologic state influencing decision making through reactive processes (Figner, Mackinlay, Wilkening, & Weber, 2009; Reyna & Farley, 2006). Nonetheless, TPB is useful in explaining the deliberative component of decision-making intentions and behaviors.

Within TPB, an individual’s intentions to perform a given behavior are directly influenced by their appraisal of the consequences of performing that behavior, and how strongly they value the presence or absence of those consequences (Ajzen, 1991). Perceived consequences of bystander intervention might be positive or negative. For example, a benefit of intervention might be having team members avoid breaking laws or team rules where disciplinary action could affect the individual or team’s eligibility for competition or success on the field of play. Conversely, athletes who believe that teammates would look unfavorably on intervention might see bystander intervention as carrying a cost. The balance of these perceived consequences influences an individual’s appraisal of the relative benefit of intervening as a bystander (Latane & Darley, 1970).

Evidence in the popular media suggests that codes of conduct and consequences for violating codes of conduct are not consistent across all colleges (Marquette Lawyer, 2012). Benford (2007) has speculated that being shielded from prosecution can contribute to a sense of entitlement among professional athletes. No research to date has evaluated empirically whether this is true among collegiate student-athletes; however, the increase in entitlement among college students overall is a topic about which there is a growing body of evidence (e.g., Greenberger, Lessard, Chen, & Farruggia, 2008; Twenge & Foster., 2010). To the extent entitlement is greater among college athletes than among their nonathlete peers, it is hypothesized that this would be explained by greater entitlement among athletes competing in Division I as compared with athletes in the typically lower public profile Division II or III; Division I football games are frequently televised for a national audience, with some teams achieving a public profile similar to that of professional athletes. Exploitative entitlement has been associated with behaviors such as aggression and disrespect for the needs of others (Greenberger et al., 2008; Lessard, Greenberger, Chen, & Farruggia, 2011; Reidy, Zeicher, Foster, & Martinez, 2008; Wink, 1991), and has been implicated in the perpetration of sexual violence (Benford, 2007; Dyson & Flood, 2008; Welch, 1997). In the context of individuals who believe they have been unfairly wronged, higher entitlement has been associated with a reduced likelihood of engaging in helping behaviors (Zitek, Jordan, Monin, & Leach, 2010). Based on this evidence, it is possible that entitlement mediates the association between perceived disciplinary consequences related to sexual violence and likelihood of intervening to prevent sexual violence or sexual behaviors that are considered to be inappropriate—however, this has not been tested empirically.

Coaches have been identified as important source of messaging for student-athletes about what constitutes appropriate and inappropriate off-field behavior (Jaime et al., 2014). No research to date has assessed whether coach expectations about appropriate off-field conduct and team consequences for violating those expectations are related to the likelihood team members will intervene to prevent inappropriate sexual behavior. However, prior evidence indicates that coaches have the capacity to positively affect the nonsport personal development of team members (Gould, Collins, Lauer, & Chung, 2007), and that relational support from coaches increases the likelihood of prosocial off-field behavior (Rutten et al., 2008).

The present study aimed to test the fit of a model predicting the likelihood of intending to intervene to prevent inappropriate sexual behavior, depicted in Figure 1. Specific model hypotheses were that (a) communication from the team coach about expectations for off-field conduct, and a belief that the team coach would strongly discipline his team for poor off-field conduct would be independently associated with intending to intervene to stop a teammate from engaging in inappropriate sexual behavior, and (b) these direct associations would be partially mediated by entitlement and attitudes about intervening. Invariance of the model by division of competition was also tested, with the hypothesis that entitlement would be greater among Division I athletes as compared with Division II or III athletes.

Figure 1.

Model for likelihood of intervening to prevent others from engaging inappropriate sexual behavior.

Method

Sample and Protocol

Data were collected in 2012 as part of the National Collegiate Athletic Association (NCAA)–sponsored National Study of Student-Athlete Social Environments. Each NCAA-member college is required to designate a faculty member to serve as a liaison between the athletics department and the university. This faculty athletics representative at each of over 1,000 NCAA-member colleges was invited to proctor a paper survey to one to three prespecified teams (selected by NCAA researchers via stratified random sampling) on their campus. Administration was standardized, but all data were anonymous at the student and school levels. Approximately 23,000 male and female collegiate student-athletes took the survey, with a school-level participation rate estimated at 65%. Full details on the broader study are available on the NCAA’s website (www.ncaa.org/research). The NCAA Research Review Board reviewed the study, as did the institutional review boards at all participating colleges. The present analyses were restricted to male football players competing in NCAA-sponsored competition (N = 3,281). A single sport was selected for this study so as to limit the possibility of confounding by characteristics related to sport participation. Football was chosen as it is the high-profile male sport with the largest number of team members on U.S. college campuses given the typical squad size.

Measures

The Social Environments survey was created by NCAA research staff, in consultation with outside experts, using or building from, where available, existing validated scales as described below. Measures used for this study are listed in full in the appendix.

Exploitative entitlement

Exploitative entitlement was measured using a four-item modification of Lessard et al.’s (2011) exploitative entitlement scale. Items were scored on a 6-point scale with response options ranging from strongly disagree to strongly agree and summed to create a composite score with higher scores indicating more exploitative entitlement. Internal consistency reliability in the present sample was acceptable (Cronbach’s α = .85).

Communicated expectations from coach

Participants were asked whether a coach or a member of the athletic department ever talked to their team about expectations about three topics: appropriate treatment of members of the opposite sex, relationship violence, and speaking up when they see things around them that are not right. Responses for each item were binary: yes (1) or no (0). The three items were summed to create a scale with higher scores indicating that more expectations were discussed. Internal consistency reliability of the measure in this sample was acceptable (Cronbach’s α = .79).

Team discipline

Expected likelihood of coach discipline was measured with responses to the item: “My coach would strongly discipline my teammates and I for poor off-field behavior.” The item was scored on a 6-point scale, with response options ranging from strongly disagree to strongly agree, and higher scores indicating a greater expectation of strong discipline for poor off-field behavior.

Attitudes about intervening

Perceived attitudes about the importance and consequences of intervening to help somebody who is in trouble were assessed with a modified version of the Banyard et al.’s (2007) Decisional Balance Scale. The scale had eight items and was scored on a 6-point scale, with response options ranging from strongly disagree to strongly agree. Items summed to create a scale with higher scores indicating more positive attitudes about intervening as a bystander. Internal consistency reliability in the present sample was acceptable (Cronbach’s α = .85).

Bystander intervention intentions

Intention of intervening to prevent inappropriate sexual behavior was assessed with a single item modified from Banyard et al.’s (2005) Bystander Willingness to Help Scale. The item asked how likely the respondent would be to “intervene in a situation to that could lead to inappropriate sexual behavior,” presuming such a situation arose. The item was scored on a 5-point scale, with response options ranging from not at all likely to extremely likely, and higher scores indicating a greater likelihood of intervening.

Division of competition

The participant’s team’s division of NCAA competition was also recorded. NCAA competition is organized into Divisions I, II, and III based on institutional factors, including the number of sports sponsored and scholarship dollars allotted. Typically, Division I is the most competitive and Division III the least competitive; however, there is variability at the team level.

Analysis

Descriptive statistics were calculated for all model variables, both for the sample as a whole and separately by division of competition. One-way ANOVA was used to compare each variable by division of competition, with Bonferroni corrections for multiple comparisons. Bonferroni-corrected Pearson pairwise correlations between all model variables were also reported. Path analysis was used to test the model presented in Figure 1. This model was overidentified with one degree of freedom. Based on the ANOVA results indicating between-division differences in several model variables, a multigroup model was also fit, with groups defined by division of competition to test for invariance of model parameters across divisions. A total of 797 participants were missing at least one model variable. Of these, 481 were excluded from the analysis because they were missing all model variables. Full information maximum likelihood estimation (STATA 12.1) was used to impute missing values for 316 cases. This method yields unbiased values under the assumption that the data are missing at random (Enders & Bandalos, 2001). Overall model fit was assessed using descriptive goodness of fit indices, with the following thresholds for close model fit: comparative fit index (CFI) > 0.95, root mean square error of approximation (RMSEA < 0.06, and standardized root mean square residual (SRMR) < 0.08 (Hu & Bentler, 1999). Analyses were conducted using STATA 12.1 (College Station, Texas). A standard alpha level of .05 was used as a threshold for significance in all analyses.

Results

The sample was comprised of 34% freshmen, 29% sophomore, 25% juniors, 10% seniors, and 3% graduate students. Over half of the sample identified as White (63.8%), 33.5% as Black, 5.1% as Hispanic, 2.1% as American Indian, 1.5% as Hawaiian or Pacific Islander, and 2.9% as Other. By division, the composition of the sample was 37% Division I, 29% Division II, and 33% Division III, relatively similar to the national breakdown of NCAA football participation by division of competition (Division I 39%, Division II 26%, Division III 35%; NCAA, 2013). Detailed descriptive statistics for all model variables are presented in Table 1. Significant differences by division in model variables were determined by a Bonferroni-corrected alpha level of less than .05. While there was significant between-division variability in attitudes about intervening, communicated expectations, and anticipated discipline, there were not significant between-division differences in exploitative entitlement or intentions of intervening.

Table 1.

Descriptive Statistics for Study Variables Among U.S. College Football Players and One-Way ANOVA for Differences in Each Variable by Division of Competition.

Variable	AllM (SD) n	Division IM (SD) n	Division IIM (SD) n	Division IIIM (SD) n	p > F
Exploitative entitlement	10.86 (4.94) 3,066	11.01 (4.92) 1,090	10.82 (5.09) 919	10.73 (4.83) 1,057	.403
Attitudes about intervening	35.38 (6.75) 2,661	35.10 (7.52) 963	34.21 (7.92) 794	35.04 (7.01) 904	.026
Communicated expectations	2.46 (0.96) 2,808	2.56 (0.91) 1,013	2.37 (1.03) 829	2.43 (0.94) 966	<.001
Anticipated discipline	4.92 (1.22) 2,975	5.05 (1.19) 1,058	4.90 (1.24) 886	5.00 (1.13) 1,031	.015
Intervene to prevent inappropriate sexual behavior	3.69 (1.34) 2,736	3.65 (1.35) 987	3.68 (1.37) 812	3.75 (1.31) 937	.197

Pairwise Pearson correlations of all measured model variables are presented in Table 2, and significance was similarly determined by a Bonferroni-corrected alpha level of less .05. As hypothesized, higher exploitative entitlement was significantly correlated with lower likelihood of intending to intervene to prevent inappropriate sexual behavior. Other hypothesized bivariate associations were also significant, with greater anticipated coach discipline for poor off-field behavior and more expectations communicated about appropriate off-field conduct associated with greater likelihood of intending to intervene to prevent inappropriate sexual behavior.

Table 2.

Pairwise Correlations Between Variables Among U.S. College Football Players.

Observed Variable	1	2	3	4	5
1. Exploitative entitlement	1.000
2. Attitude about intervening	.07*	1.000
2. Attitude about intervening	n = 2,549
3. Communicated expectations	−.15*	.21*	1.000
3. Communicated expectations	n = 2,670	n = 2,600
4. Anticipated discipline	−.15*	.21*	.25*	1.000
4. Anticipated discipline	n = 2,860	n = 2,568	n = 2,702
5. Intervene	−.06*	.26*	.19*	.13*	1.000
	n = 2,626	n = 2,605	n = 2,689	n = 2,651

Note. Communicated expectations = communicated expectations from coach or athletic department for off-field conduct; Anticipated discipline = anticipated strong discipline from coach for team’s poor off-field behavior; Intervene = likelihood of intervening to prevent inappropriate sexual behavior. Bonferroni adjustment made for multiple comparisons.

p < .05.

Direct, indirect, and total effects as well as fit statistics for the path analyses model for the full sample are reported in Table 3, with standardized path coefficients for the full sample and by division of competition shown in Figure 2. The nearly fully saturated model had fit statistics that indicated a close model fit (Hu & Bentler, 1999). Thirteen percent (13.3%) of the variation in the dependent variables is explained by the model, comprised of 4.1% of the variation in exploitative entitlement, 8.3% of the variation in attitudes about intervening and 9.3% of the variation in likelihood of intending to intervene to prevent inappropriate sexual behavior.

Table 3.

Direct and Indirect Effects in Model Predicting Intervening to Prevent Inappropriate Sexual Behavior in a Sample of College Football Players (N = 3,281).

Causal Variable	Endogenous Variables
	Entitlement			Attitude About Intervening			Intervening
	Unstandardized	SE	Standardized	Unstandardized	SE	Standardized	Unstandardized	SE	Standardized
Anticipated discipline
Direct effect	−0.49*	0.08	−0.12	1.13*	0.13	0.18	0.05*	0.02	0.05
Total indirect effects	—	—	—	−0.10*	0.02	−0.02	0.05*	0.01	0.04
Total effects	−0.49*	0.08	−0.12	1.04*	0.13	0.17	0.10*	0.02	0.09
Communicated expectations
Direct effect	−0.63*	0.10	−0.12	1.49*	0.15	0.19	0.18*	0.03	0.13
Total indirect effects	—	—	—	−0.12*	0.03	−0.02	0.06*	0.01	0.05
Total effects	−0.63	0.10	−0.12	1.37*	0.15	0.18	0.24*	0.03	0.17
Entitlement
Direct effect	—	—	—	0.20*	0.03	0.13	−0.01*	0.01	−0.05
Total indirect effects	—	—	—	—	—	—	0.01*	0.00	0.03
Total effects	—	—	—	0.20*	0.03	0.13	−0.00	0.00	−0.02
Attitude about intervening
Direct effect	—	—	—	—	—	—	0.04*	0.00	0.22
Total indirect effects	—	—	—	—	—	—	—	—	—
Total effects	—	—	—	—	—	—	0.04*	0.00	0.22

Note. Fit statistics: CFI = 1.00, RMSEA < 0.001, SRMR < 0.001. CFI =comparative fit index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual.

p < .05.

Figure 2.

Path model showing standardized direct effects for full sample and by division of competition of the likelihood of U.S. college football players intervening to prevent others from engaging inappropriate sexual behavior (N = 3,281).

We hypothesized that entitlement would mediate the association between anticipated coach discipline and likelihood of intending to intervene to prevent inappropriate sexual behavior and between communicated expectations and likelihood of intending to intervene. As presented in Table 3, the total indirect relationship between anticipated coach discipline and likelihood of intervening (β = .04) was significant for an alpha level of .05, as was the total indirect relationship between communicated expectations and likelihood of intending to intervene (β = .05). Attitudes about the consequences of intervening significantly mediated the association between expected discipline, communicated expectations, and entitlement and likelihood of intending to intervene to prevent inappropriate sexual behavior. The indirect pathway between expected coach discipline and likelihood of intending to intervene, mediated through entitlement accounted, for 6.2% of the variation between expected coach discipline and likelihood of intending to intervene. The indirect pathway between these two variables through the athlete’s attitude about intervening accounted for 45.2% of the total association. The indirect pathway between communicated coach expectations and likelihood of intending to intervene, mediated through entitlement, accounted for 3.2% of the association between these two variables, whereas the indirect association between communicated coach expectations and likelihood of intending to intervene, mediated through the athlete’s attitude about intervening, accounted for 24.1% of the total association.

We allowed the model parameters to vary freely across divisions of competition and used a Wald test to test for invariance of structural parameters across these three groups. Although overall the joint test for differences in structural parameters between models was not significant, χ²(18) = 22.33, p > χ² = 0.22, we found that the path from expected discipline to entitlement was significantly weaker in Division I than Divisions II and III, χ²(2) = 9.74, p > χ² = 0.008. When the model was restricted just to Division I athletes, we found the indirect paths from discipline and communicated expectations to likelihood of intending to intervene through entitlement and the direct path between expected discipline and likelihood of intending to intervene to be nonsignificant. Among Division I athletes, coach expectations for off-field conduct were associated with bystander intervention intentions both directly and indirectly through the athlete’s attitudes about intervening.

Discussion

Reducing the incidence of sexual violence on U.S. college campuses is an important and necessary task, and bystander-approaches have been identified as a useful component of prevention. Identifying contextual factors that facilitate or constrain the likelihood that specific groups of bystanders intervene to prevent sexual violence is important if we wish to create contexts that encourage intervention. So far, most research about contextual influences on bystander behavior has focused on the role of the peer environment. The present findings reveal some other potentially important aspects within the athletic environment.

It is important to note that outcome measured in this study was “inappropriate sexual behavior,” a construct that is inherently subjective. In the sample as a whole, communicated expectations about and expected discipline from the team coach for poor off-field conduct were associated with an individual’s likelihood of intending to intervene to prevent what they consider to be inappropriate sexual behavior, both directly and indirectly mediated by the individual’s attitudes about the consequences and importance of intervening and his level of exploitative entitlement. The indirect pathway through attitudes about intervening is consistent with Latane and Darley’s (1970) rational stages for deciding whether or not to intervene. These stages include individuals’ assessment that a situation is problematic, that they are responsible for taking action, and their appraisal that benefits of intervening outweigh the costs. Communicated expectations from the team coach or athletic department about relationship violence, appropriate treatment of the opposite sex, and the importance of taking action “if you see things that aren’t right,” set injunctive norms that may help shift an individual’s assessment that a current situation is problematic and that he or she has responsibility to take action. Expectations of “strong discipline” from the team coach for poor off-field behavior reflect a potential cost of not intervening. Ajzen’s (1991) TPB suggests that population-salient consequences of performing a behavior are predictive of whether an individual performs that behavior. The significant association between expected discipline for poor off-field behavior and the individual’s attitudes about the consequences of and importance of intervening suggests that anticipating coach discipline may be one salient motivation for intervening. However, only 8.3% of the variation in attitudes about the consequences and importance of intervening were explained by the model tested. This suggests that there are other important factors beyond those included in the present analysis. Other important factors could include perceived team norms about intervening and beliefs about what constitute inappropriate sexual behavior. Additional research is needed to identify other modifiable individual and contextual factors that change the rational cost–benefit calculation about intervening to prevent inappropriate sexual behavior.

The other indirect pathway through which communicated expectations and expected discipline were associated with intentions of intervening was through exploitative entitlement. Benford (2007) has suggested that among professional athletes, a perception that they are less likely to be prosecuted for crimes contributes to a sense of entitlement. The study finds an association between expectations about discipline and entitlement among Division II and III athletes, but not among Division I athletes. This association is cross-sectional and does not indicate whether there exists a causal relationship between these two variables. Additional research is needed to clarify what factors predict variability in entitlement among Division I football players, a question that is particularly interesting as the mean entitlement scores were not significantly different by division of competition. Among athletes in all divisions, those with higher levels of exploitative entitlement indicated that they would be less likely to intervene to prevent what they consider to be inappropriate sexual behavior. It is critical, however, to note that the results of this study do not indicate that collegiate football players have higher levels of exploitative entitlement than their nonathlete peers, but rather that variability in exploitative entitlement is one mediating pathway through which coach expectations and discipline may be related to bystander behavior. Notably, the mean entitlement item scores for the football players in the present sample (2.72 out of 6, SE = 1.24) were nearly identical to those observed by Lessard et al.’s (2011) in a general undergraduate sample (2.71, SE = 0.72) and fewer than 4% of the football players reported high entitlement scores of greater than or equal to 20 out of 24.

Across the full sample, the association between exploitative entitlement and intentions about intervening is partially mediated by the individual’s attitudes about the consequences and importance of intervening in general; interestingly, individuals with a high level of exploitative entitlement were more likely to hold positive attitudes about intervening. It is possible that responses to some of the questions that composed the attitudes measure, such as “Teammates will look up to and admire me if I intervene,” could be correlated with the constellation of personality traits related to entitlement such as narcissism. Nonetheless, given the direct pathway indicating the higher entitlement is associated with lower likelihood of intending to intervene to prevent inappropriate sexual behavior, it is not a benign construct. Additional research is needed to explore the negative and potentially positive dimension of exploitative entitlement as it relates to bystander behaviors.

Ensuring that expectations for off-field behavior are set and consequences for violating these expectations are enforced may be important components of creating an athletic environment that encourages bystander intervention by modifying reinforcements and the cost–benefit appraisal of intervening. Team coaches are well positioned to take on the responsibility for setting and enforcing expectations for off-field conduct. However, not all coaches may be comfortable or motivated to do so. Athletic departments may thus need to take responsibility for providing a framework within which coaches are informed about the importance of setting and enforcing standards for off-field behavior, and appropriately incentivized to do so. This might involve setting department-wide expectations about coach–athlete communication about off-field conduct, or holding coaches accountable for the off-field conduct of team members. As part of this process, it may be useful to provide coaches with tools and strategies for communicating with athletes about these topics.

Limitations

There are several important limitations to this study. Most centrally, the cross-sectional design limits our ability to make causal inferences about the model. Future longitudinal research is necessary to better address issues of causality. In addition, the nearly fully saturated nature of the model tested introduces the possibility that there are equivalent models that fit as well as the one hypothesized and tested.

There are limitations related to the measures used. Expected coach discipline was measured using a single item and information is not available about the reliability or validity of that item. In addition, the construct as measured is inherently subjective. Different coaches may engage in different types of disciplinary behaviors, and different respondents could have varying conceptions of what it means to be “strongly disciplined” by a team coach. Consequently, the results of this study do not provide information about the type of disciplinary action that motivates bystander behavior. In addition, the measure asks about expectations of strong discipline for “off-field behavior.” This could include a range of behaviors other than, or in addition to, inappropriate sexual behavior. It is recommended that future research in this area assess specific off-field behaviors, as opposed to the broader class of off-field behaviors. Research is also needed to determine the types of disciplinary measures that motivate athlete compliance—for example, loss of playing time.

The primary dependent variable asks participants how likely they would be to intervene to prevent inappropriate sexual behavior. Presumably, responses to this question reflect both intention and an appraisal of barriers to engaging in this action. However, this is not the same as recording actual behavior. The wording of the outcome measure itself is a limitation. It asks about “inappropriate sexual behavior”; however, the way it is worded makes this a subjective construct. Thus, responses to this question reflect both an appraisal of what are in fact inappropriate sexual behaviors, and the individual’s likelihood of intervening should a situation meeting this definition arise. Sexual behaviors other than sexual violence that are considered to be inappropriate, perhaps referring to the characteristics of the individual at whom the sexual behavior is directed, may be subsumed within this measure. Consequently, there may be between-individual variation in responses to this measure that reflect something other than the likelihood of intervening. Future research may consider more clearly specifying the types of inappropriate behavior for which an individual would be likely to intervene. A further limitation of measuring intervention intentions is that while intentions are often strongly predictive of behavior (Ajzen, 1991), they are not a proxy for behavior. The association between intentions and behavior presumes a deliberative decision-making process. However, decision making is often reactive and nondeliberative, and this type of decision making, in which intentions are less strongly related to behavior, is more likely to occur under conditions of strong emotions or physiologic arousal (Figner et al., 2009; Reyna & Farley, 2006). Consequently, while participants in the present study may truly believe that they would engage as a prosocial bystander, when they are out socializing with teammates their decisional calculus may look very different than when they are completing a survey about their intentions .

Finally, the communicated expectations measure asks whether any communication about specific topics occurred, but does not assess the nature or frequency of this communication. Future research including qualitative interviews with athletes would be useful to provide a more complete understanding of coach–athlete communication about these topics.

There may be limitations related to the sample used and the generalizability of the results. The present study looked only at male football players at U.S. colleges. The results may not generalize to other sports, ages, levels of competition, or female athletes. In addition, it is not clear whether the model studied would be relevant to nonsport contexts. College sports are unique in that there are well-established enforcement and eligibility mechanisms: coach, athletic department, institution, conference, and national governing bodies (e.g., NCAA, National Association of Intercollegiate Athletics, and National Junior College Athletic Association). A team’s success and an individual’s continued eligibility to participate on his or her team may be powerful motivators for compliance with expected codes of conduct; there may not be similarly salient consequences of violating codes of conduct, or enforcement mechanisms for codes of conduct, in other high-risk groups (e.g., fraternity members). There may even be team selection bias in that coaches with stringent notions of proper off-field conduct may select team members based on prior off-field behavior.

Conclusion

Athletes can play an important role in the prevention of sexual violence by intervening to prevent teammates and other peers from perpetrating these behaviors. To create athletic environments that encourage such bystander behaviors, it is important to understand what factors make it more or less likely that an athlete will intervene. The present findings indicate that when coaches and athletic departments clarify expectations for off-field conduct and strongly sanction violations of this conduct they may be indirectly encouraging team members to engage in bystander intervention by influencing the calculation of the costs and benefits of intervening. They may also be reducing exploitative entitlement among athletes, lower levels of which are associated with a greater likelihood of intervening. Coaches and athletic departments are encouraged to support ongoing psycho-educational programming for athletes about bystander behaviors by providing a context in which coaches are encouraged and supported to set and enforce behavioral standards for their student-athletes.

Footnotes

Appendix

Authors’ Note

Work on this project was completed while Emily Kroshus was a Post-Doctoral Research Fellow at the NCAA Sport Science Institute.

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) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Emily Kroshus is a research assistant professor at University of Washington in the Department of Pediatrics. She is also a consultant for the National Collegiate Athletic Association (NCAA) Sport Science Institute, having previously been employed by the NCAA as a postdoctoral research fellow. She received her ScD from the Harvard School of Public Health in Social and Behavioral Sciences, her MPH from Johns Hopkins School of Public Health, and a BA in economics from Princeton University. She is broadly interested in translational health communication research as it relates to health and safety in sport and exercise settings, including understanding individual, interpersonal, and contextual determinants of relevant primary and secondary preventive behaviors and developing and evaluating interventions to modify these behaviors.

Tom Paskus is the principal research scientist for the NCAA. In this role, he directs the NCAA’s national portfolio of studies on the academic trajectories of college student-athletes and oversees the NCAA’s data collections and research initiatives pertaining to the academic, athletic, social, and personal well-being of current and former student-athletes. He has worked extensively on the development of the Division I Academic Progress Rate (APR), the Division I and II initial eligibility and progress toward degree standards, the Growth, Opportunities, Aspirations and Learning of Students in college (GOALS) and Study of College Outcomes and Recent Experiences (SCORE) studies of the college student-athlete experience, and the NCAA sports wagering study. Prior to joining the NCAA, he was a faculty member in the Quantitative Research Methods Program in the College of Education at the University of Denver. He received his PhD and MA in quantitative psychology from the University of Virginia, and an AB in psychology from Dartmouth College.

Lydia Bell is the associate director of Research for Academic Performance at the NCAA. In this role, she assists in all aspects of development and analysis of research on current and former student-athlete academic performance and well-being. She also oversees the Innovations in Research and Practice Grant Program, and frequently collaborated with the Sport Science Institute on their student-athlete well-being and mental health initiatives. Prior to joining the NCAA, she was an assistant professor of practice and the director of Project Student Outread for Access & Resiliency (SOAR) in the Center for the Study of Higher Education at the University of Arizona. She received her PhD in language, reading, and culture and MA in higher education from the University of Arizona, and an AB in government and legal studies and sociology from Bowdoin College. She served as the chair of the American Educational Research Association’s SIG Research Focus on Education and Sport from 2010 to 2013 and is on the editorial board of the Journal for the Study of Sports and Athletes in Education.

References

Ajzen

(1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179-211.

Armstrong

E. A.

Hamilton

Sweeney

(2006). Sexual assault on campus: A multilevel, integrative approach to party rape. Social Problems, 53, 483-499.

Banyard

V. L.

(2011). Who will help prevent sexual violence: Creating an ecological model of bystander intervention. Psychology of Violence, 1, 216-229.

Banyard

V. L.

Cross

Modecki

K. L.

(2006). Interpersonal violence in adolescence ecological correlates of self-reported perpetration. Journal of Interpersonal Violence, 21, 1314-1332.

Banyard

V. L.

Moynihan

M. M.

(2011). Variation in bystander behavior related to sexual and intimate partner violence prevention: Correlates in a sample of college students. Psychology of Violence, 1, 287-301.

Banyard

V. L.

Moynihan

M. M.

Crossman

M. T.

(2009). Reducing sexual violence on campus: The role of student leaders as empowered bystanders. Journal of College Student Development, 50, 446-457.

Banyard

V. L.

Moynihan

M. M.

Plante

E. G.

(2007). Sexual violence prevention through bystander education: An experimental evaluation. Journal of Community Psychology, 35, 463-481.

Banyard

V. L.

Plante

E. G.

Moynihan

M. M.

(2004). Bystander education: Bringing a broader community perspective to sexual violence prevention. Journal of Community Psychology, 32, 61-79.

Banyard

V. L.

Plante

E. G.

Moynihan

M. M.

(2005). Rape prevention through bystander education (Final Report to NIJ for Grant 2002-WG-BX-0009). Retrieved from http://www.ncjrs.gov/pdffiles1/nij/grants/208701.pdf

10.

Banyard

V. L.

Ward

Cohn

E. S.

Plante

E. G.

Moorhead

Walsh

(2007). Unwanted sexual contact on campus: A comparison of women’s and men’s experiences. Violence and Victims, 22, 52-70.

11.

Bénabou

Tirole

(2004). Incentives and prosocial behavior (Discussion Papers in Economics No. 230). Woodrow Wilson School of Public and International Affairs, Princeton University. Retrieved from http://www.cepr.org/active/publications/discussion_papers/dp.php?dpno=4633

12.

Benford

R. D.

(2007). The college sports reform movement: Reframing the “edutainment” industry. The Sociological Quarterly, 48, 1-28.

13.

Boeringer

S. B.

(1999). Associations of rape-supportive attitudes with fraternity and athletic participation. Violence Against Women, 5, 81-90.

14.

Bronfenbrenner

(1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.

15.

Brown

Messman-Moore

T. L.

(2010). Personal and perceived peer attitudes supporting sexual aggression as predictors of male college students’ willingness to intervene against sexual aggression. Journal of Interpersonal Violence, 25, 503-517.

16.

Burn

S. M.

(2009). A situational model of sexual assault prevention through bystander intervention. Sex Roles, 60, 779-792.

17.

Casey

E. A.

Lindhorst

T. P.

(2009). Toward a multi-level, ecological approach to the primary prevention of sexual assault prevention in peer and community contexts. Trauma, Violence, & Abuse, 10, 91-114.

18.

Chandler

S. B.

Dewayne

J. J.

Carroll

P. S.

(1999). Abusive behaviors of college athletes. College Student Journal, 33, 638-645.

19.

Chaurand

Brauer

(2008). What determines social control? People’s reactions to counternormative behaviors in urban environment. Journal of Applied Social Psychology, 38, 1689-1715.

20.

Cleere

Lynn

S. J.

(2013). Acknowledged Versus Unacknowledged Sexual Assault Among College Women. Journal of Interpersonal Violence, 28(12), 2593-2611.

21.

Cole

T. B.

(2006). Rape at US colleges often fueled by alcohol. Journal of the American Medical Association, 296, 499-505.

22.

Dyson

Flood

(2008). Building cultures of respect and non-violence. Retrieved from http://www.2shine.org.nz/library/Documents/RWC%20special%20collection/respectresponsibilityreport.ashx.pdf

23.

Enders

C. K.

Bandalos

D. L.

(2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling, 8, 430-457.

24.

England

Thomas

R. J.

(2006). The decline of the date and the rise of the college hook up. The Family in Transition, 14, 151-162.

25.

Figner

Mackinlay

R. J.

Wilkening

Weber

E. U.

(2009). Affective and deliberative processes in risky choice: Age differences in risk taking in the Columbia Card Task. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35, 709-730.

26.

Gidycz

C. A.

Orchowski

L. M.

Berkowitz

A. D.

(2011). Preventing sexual aggression among college men: An evaluation of a social norms and bystander intervention program. Violence against Women, 17, 720-742.

27.

Gould

Collins

Lauer

Chung

(2007). Coaching life skills through football: A study of award winning high school coaches. Journal of Applied Sport Psychology, 19(1), 16-37.

28.

Greenberger

Lessard

Chen

Farruggia

S. P.

(2008). Self-entitled college students: Contributions of personality, parenting and motivational factors. Journal of Youth and Adolescence, 37, 1193-1204.

29.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

30.

Jaime

M. C. D.

McCauley

H. L.

Tancredi

D. J.

Nettiksimmons

Decker

M. R.

Silverman

J. G.

Miller

(2014). Athletic coaches as violence prevention advocates. Journal of Interpersonal Violence, 30, 1090-1111. doi:10.1177/0886260514539847

31.

Koelsch

L. E.

Brown

A. L.

Boisen

(2012). Bystander perceptions: Implications for university sexual assault prevention programs. Violence and Victims, 27, 563-579.

32.

Latane

Darley

(1970). The unresponsive bystander: Why doesn’t he help? Englewood Cliffs, NJ: Prentice Hall.

33.

Lessard

Greenberger

Chen

Farruggia

(2011). Are youths’ feelings of entitlement always “bad”? Evidence for a distinction between exploitive and non-exploitive dimensions of entitlement. Journal of Adolescence, 34, 521-529.

34.

Marquette Lawyer . (2012, Summer). Penalties and procedures: Increasingly complex off-the-field rules and processes for athletes mean more interaction between the worlds of sports and law (pp. 16-21). Retrieved from http://law.marquette.edu/assets/marquette-lawyers/pdf/marquette-lawyer/2012-summer/2012-summer-p16.pdf

35.

McCauley

Ruggiero

K. J.

Resnick

H. S.

Conoscenti

L. M.

Kilpatrick

D. G.

(2009). Forcible, drug-facilitated, and incapacitated rape in relation to substance use problems: Results from a national sample of college women. Addictive Behaviors, 34, 458-462.

36.

McMahon

Banyard

V. L.

(2012). When can I help? A conceptual framework for the prevention of sexual violence through bystander intervention. Trauma, Violence, & Abuse, 13, 3-14.

37.

Messman-Moore

T. L.

Coates

A. A.

Gaffey

K. J.

Johnson

C. F.

(2008). Sexuality, substance use, and susceptibility to victimization: Risk for rape and sexual coercion in a prospective study of college women. Journal of Interpersonal Violence, 23, 1730-1746.

38.

Miller

Tancredi

D. J.

McCauley

H. L.

Cecker

M. R.

Virata

M. C. D.

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(1), 108-112.

39.

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.

40.

Mouilso

E. R.

Fischer

(2012). A prospective study of sexual assault and alcohol use among first-year college women. Violence and Victims, 27, 78-94.

41.

Moynihan

M. M.

Banyard

V. L.

(2008). Community responsibility for preventing sexual violence: A pilot study with campus Greeks and intercollegiate athletes. Journal of Prevention & Intervention in the Community, 36(1-2), 23-38.

42.

Moynihan

M. M.

Banyard

V. L.

(2011). Educating bystanders helps prevent sexual violence and reduce backlash. Family & Intimate Partner Violence Quarterly, 3, 293-303.

43.

Moynihan

M. M.

Banyard

V. L.

Arnold

J. S.

Eckstein

R. P.

Stapleton

J. G.

(2010). Engaging intercollegiate athletes in preventing and intervening in sexual and intimate partner violence. Journal of American College Health, 59, 197-204.

44.

Murnen

S. K.

Kohlman

M. H.

(2007). Athletic participation, fraternity membership, and sexual aggression among college men: A meta-analytic review. Sex Roles, 57, 145-157.

45.

National Collegiate Athletic Association. (2013). NCAA sports sponsorship and participation rates report. Retrieved from http://www.ncaapublications.com/productdownloads/PR2014.pdf

46.

Palmer

R. S.

McMahon

T. J.

Rounsaville

B. J.

Ball

S. A.

(2010). Coercive sexual experiences, protective behavioral strategies, alcohol expectancies and consumption among male and female college students. Journal of Interpersonal Violence, 25, 1563-1578.

47.

Marquette Lawyer . (2012, Summer). Penalties and procedures: Increasingly complex off-the-field rules and processes for athletes mean more interaction between the worlds of sports and law. Retrieved from http://law.marquette.edu/assets/marquette-lawyers/pdf/marquette-lawyer/2012-summer/2012-summer-p16.pdf

48.

Reidy

D. E.

Zeicher

Foster

J. D.

Martinez

M. C.

(2008). Effects of narcissistic entitlement and exploitativeness on human physical aggression. Personality and Individual Differences, 44, 865-875.

49.

Reyna

V. F.

Farley

(2006). Risk and rationality in adolescent decision making: Implications for theory, practice, and public policy. Psychological Science in the Public Interest, 7(1), 1-44.

50.

Rutten

E. A.

Deković

Stams

G. J. J.

Schuengel

Hoeksma

J. B.

Biesta

G. J.

(2008). On- and off-field antisocial and prosocial behavior in adolescent soccer players: A multilevel study. Journal of Adolescence, 31, 371-387.

51.

Sawyer

R. G.

Thompson

E. E.

Chicorelli

A. M.

(2002). Rape myth acceptance among intercollegiate student athletes: A preliminary examination. American Journal of Health Studies, 18(1), 19-25.

52.

Twenge

J. M.

Foster

J. D.

(2010). Birth cohort increases in narcissistic personality traits among American college students, 1982-2009. Social Psychological and Personality Science, 1, 99-106.

53.

Vladutiu

C. J.

Martin

S. L.

Macy

R. J.

(2011). College- or university-based sexual assault prevention programs: A review of program outcomes, characteristics, and recommendations. Trauma, Violence, & Abuse, 12(2), 67-86.

54.

Webb

T. L.

Sheeran

(2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological bulletin, 132(2), 249-268.

55.

Welch

(1997). Violence against women by professional football players. A gender analysis of hypermasculinity, positional status, narcissism, and entitlement. Journal of Sport & Social Issues, 21, 392-411.

56.

Wink

(1991). Two faces of narcissism. Journal of Personality and Social Psychology, 61, 590-597.

57.

Zitek

E. M.

Jordan

A. H.

Monin

Leach

F. R.

(2010). Victim entitlement to behave selfishly. Journal of Personality and Social Psychology, 98, 245-255.