Sexual harassment continues to be a significant problem for women in the U.S. workforce. Despite refinements to sexual harassment law over the decades, and the rapidly changing gender composition of the workforce, the prevalence of women experiencing sexually-harassing behaviors at work has remained consistently high in recent years, with the majority of working women experiencing at least one incident of harassment (Berdahl and Aquino 2009; Cortina et al. 2013; Ilies et al. 2003; Kabat-Farr and Cortina 2014). The recent gap in the literature on organizational sexual harassment from the 1990s through most of the first decade of the 2000s, coupled with the continued prevalence of sexual harassment in organizations (Cortina et al., 2011; Kabat-Farr and Cortina 2014), have made solution-focused sexual harassment research urgent.

The consequences of this pervasive form of gender-based harassment are numerous and stark. Being the victim of sexual harassment can damage women’s prospects for gaining employment, advancing in their careers, and attaining higher wages, while also potentially creating an offensive work environment that interferes with job performance (Cortina et al. 2013; Gettman and Gelfand 2007; Miner-Rubino and Cortina 2004; Stockdale 1998). Sexual harassment can also be detrimental to the career aspirations of female employees. Women that experience sexual harassment in the workplace report being more fearful of diminished promotional opportunities, being fired from the organization, and experiencing an unfriendly work environment that may obstruct their capacity to perform essential job functions (Buchanan et al. 2014; Dougherty et al. 1996; Holland and Cortina 2016).

Sexual harassment has also proven costly to organizations. In the U.S. in 2014, $106 million in legal payments and monetary benefits to victims of sexual harassment was spent by organizations across more than 26,000 sex-based discrimination charges filed to the Equal Employment Opportunity Commission (EEOC) under Title VII. The EEOC also recovered an estimated $35 million dollars in monetary benefits associated with 6862 cases of resolved sexual harassment (EEOC 2015). Finally, millions are spent annually by organizations on lost hours and turnover associated with a harassing work environment, increasing the financial importance of preventing sexually harassing behaviors (Berkley and Kaplan 2009; Boyd 2010). Sexual harassment in the United States, and abroad (AWARE 2008; ROCU 2014; Trade Union Congress 2016), has remained a prevalent problem in recent years. Due to the unique national laws and cultural norms around sexual harassment, however, we focus the present research just on participants and organizations in the U.S.

The Influence of Organizational Policies on Sexual Harassment

For the reasons above, organizations have become increasingly concerned with reducing workplace sexual harassment and creating anti-harassment organizational cultures through the use of formal policies (Gruber 1998; Schultz 2003; Williams et al. 1999). Indeed, the federal government has advocated for the implementation of organizational policies against sexual harassment to protect women’s access to traditionally-male jobs, forbid expressions of sexist hostility, and encourage a more equitable vertical and horizontal distribution of men and women employees (Schultz 2003). Organizations have also implemented policies around sexual harassment to reduce occurrences and avoid legal liability (Stockdale et al. 2004).

Of the various organizational policies on sexual harassment in organizations, zero-tolerance policies appear to have the greatest potential to reduce harassment (Anton 2015; Gruber 1998; Magley and Shupe 2005; Offermann and Malamut 2002; Perry et al. 2010; Schultz 2003; Williams et al. 1999). For example, according to the Department of Defense 1995 sexual harassment survey (Bastian et al. 1996), the U.S. army exhibited a sharp decrease in sexual harassment prevalence after the implementation of a zero-tolerance sexual harassment policy- with 55% of women in 1995 experiencing sexual harassment compared with 64% of women in 1988. A study by Gruber (1998) also suggested that zero-tolerance policies may decrease the prevalence of sexual harassment compared to standard policies through an increased likelihood of victims reporting their victimization. However, all research to date on the relationship between sexual harassment policies and organizational outcomes has been correlational. Without experimental work it is impossible to determine if the policies themselves are responsible for observed changes in organizational outcomes. Decreases in sexual harassment observed after the implementation of a zero-tolerance policy may, for example, be due to a regression to the mean after a peak in harassment perpetration. These reductions could also be due to concurrent changes in culture, climate, and leadership, or even to changes in the measurement of sexual harassment that were implemented alongside policy changes.

Advancing Research on Influence of Organizational Policies on Sexual Harassment

The first step towards advancing research on the effects of organizational policies on sexual harassment outcomes is to perform experimental research that tests a causal relationship between harassment-related policies and harassment outcomes at work. The current work seeks to do exactly that. In developing our predictions about the conditions under which policies will impact harassment outcomes, we were guided by schematic models of social information processing (Macrae and Quadflieg 2010; Smith 1998).

Organizational Policies as Schemas

Schemas are cognitive templates that help us comprehend and respond to experiences by providing pre-organized, general-purpose understandings that can be adapted to the specifics of the current situation. Schemas are thought to serve the primary function of guiding the interpretation of new, unspecified, or ambiguous situations (Bodenhausen et al. 2003; Marshall, 2017; Valliere, 2013). We expect that company policies, like schemas, can serve as generic knowledge structures that organize a person’s processing, interpretation, and decision making related to workplace information (Bodenhausen 1992; Bodenhausen et al. 2003; Valliere, 2013). Indeed, research has shown that organizational policies can provide a basis for making inferences about unspecified elements of a situation and guide the interpretation of information that occurs in the organizational environment (Bodenhausen et al. 2003; Valliere, 2013).

Research also suggests that we are especially likely to rely on schemas when confronted with ambiguous information (Bodenhausen et al. 2003; Marshall, 2017). For example, Shotland and Straw (1976) found that when people observe an ambiguous situation in which a man is sexually harassing a woman on the street, they often assume it is a couple’s quarrel and fail to take any steps to help the female victim. Only when the couple’s quarrel schema was rendered inapplicable by the woman exclaiming to the perpetrator “I don’t know you!” did bystanders interpret the situation as requiring intervention. Conspicuous or flagrant forms of sexual harassment, such as unwanted physical advances, can also serve as cues for activating the sexual harassing behavior schema, enabling individuals to label the behavior as sexual harassment (Magley and Shupe 2005). Thus, in some situations, an appropriate schema is activated and utilized because the current situation unambiguously falls into the schema category.

However, often there is no clear schema that fits the specifics of a particular situation and/or multiple schemas are potentially applicable. Schema theory indicates that in such ambiguous situations, the schema that has the highest degree of accessibility to the perceiver is typically utilized (Bodenhausen et al. 2003; Bruner 1957). Indeed, easily accessible schemas that have been frequently and recently used are more likely to be applied to ambiguous situations than competing schemas that are not as accessible (Bem 1981; Bodenhausen et al. 2003; Bruner 1957; Valliere, 2013).

To determine whether organizational policies operate like schemas to affect sexual harassment outcomes, we needed to test whether salient sexual harassment policies exert a stronger influence on people’s perceptions of and behaviors towards sexual harassment when the harassment is ambiguous versus when it is glaring and unmistakable. Thus, we sought to examine the effects of organizational policies on both severe and more moderate or ambiguous forms of workplace harassment. Critically, this examination also expands the literature on workplace sexual harassment more broadly, as most of the existing experimental studies on sexual harassment have focused on assessing severe forms of sexual harassment, despite the fact that most incidents of sexual harassment in the workplace are ambiguous in nature (Firestone et al. 2012).

Sexual Harassment Severity

Sexual harassment severity has been defined legally and psychologically, with much overlap between the two definitions. In terms of legal characterizations, the EEOC’s Guidelines on Discrimination because of Sex (EEOC 1990) recognize “quid pro quo” harassment as one form of harassment in which submission to or rejection of unwelcome sexual conduct by an individual is used for the basis of some employment decision affecting that individual. Previous research and case law has found that quid pro quo harassment is generally regarded as one of the most severe forms of harassment by both men and women (Hames, 1994; Keyton et al. 2001; Linenberger 1983), though it occurs less frequently than other types of harassment.

The other form of sexual harassment recognized by the EEOC is “hostile environment harassment.” This term refers to when unwelcome sexual conduct “unreasonably interferes with an individual's job performance” or creates an “intimidating, hostile or offensive work environment” (EEOC 1990, 29 C.F.R. § 1604.11a). These behaviors are considered to be more ambiguous and less severe than quid pro quo harassment, and include unwanted remarks of a sexual nature, repeated requests for dates, whistles, staring, and sexual propositions not directly linked to job enhancement or job threat (Charney and Russell 1994). Previous analyses of sexual harassment complaints revealed that the most frequent forms of sexually harassing behavior filed on complaint records were mild to moderate in nature, and consisted of “unwanted physical conduct, offensive language, sexual propositions not related to employment conditions, and socialization or date requests” (Firestone et al. 2012; Popovich et al. 1992; Terpstra and Cook 1985).

The difference between quid pro quo harassment and hostile environment harassment in the legal literature is mirrored in the two main forms of sexual harassment in psychological research: sexual coercion and unwanted sexual attention, respectively (Gelfand et al. 1995). Much like quid pro quo harassment, sexual coercion includes behaviors that either overtly or implicitly link sexual compliance to job outcomes (Tata 2000). Unwanted sexual attention, which includes behaviors initiated for the purpose of gaining sexual cooperation or advances that are not welcomed by the target or related to job outcomes, falls within the EEOC’s definition of “hostile environment harassment.”(Equal Employment Opportunity Commission,1980).

Consistent with legal definitions and precedents, psychological research has determined that sexual coercion is more likely to be perceived as harassing and severe than unwanted sexual attention by college students (Fitzgerald and Ormerod 1991; Terpstra and Baker 1989) and psychological researchers and practitioners (Berkley and Kaplan 2009; Fitzgerald and Ormerod 1991; Fitzgerald et al. 1997; Kaser et al. 1995; Keyton et al. 2001; Linenberger 1983). However, while researchers generally concur on the extreme forms of sexual harassment, they do not always agree on the types of unwanted sexual attention that should be defined as sexual harassment (Osman 2007). Part of the lack of consensus regarding what types of unwanted sexual behaviors are severe enough to constitute hostile environment harassment may be due to the fact that few social science studies have empirically disentangled harassment subtypes (Kabat-Farr and Cortina 2014). Instead, the majority of recent research focuses on severe forms of harassment involving sexual-advances, while less is known about the more frequently occurring moderate forms of gender harassment that do not involve sexual advances (Cortina et al. 2013; Kabat-Farr and Cortina 2014; Leskinen et al. 2011).

Coworker Perpetrated Harassment

Finally, in addition to a narrow focus on severe sexual harassment behaviors, most previous research regarding sexual harassment has focused on the behavior of a supervisor perpetrator (Firestone et al. 2012). While the prevalence of supervisor perpetrated sexual harassment is certainly a problem, a higher percentage of women report experiencing sexually harassing behaviors from their coworkers than from their supervisors (Berdahl and Raver 2011). For example, a 2014 national survey conducted by the Restaurant Opportunities Centers United (ROCU) across major U.S. cities found that that over 55% of the employed women surveyed reported experiencing behaviors at work that constituted sexual harassment in the previous month. Additionally, of the employed women who reported experiencing behaviors that constituted sexual harassment at work in the previous month, 66% reported experiencing the harassing behaviors from management and 80% reported experiencing harassing behaviors from coworkers (ROCU 2014). Similarly, a 2008 survey conducted by the Association of Women for Action and Research (AWARE) that included 500 respondents from 92 companies found that 27% of women experienced harassment from a colleague, compared to 17% by upper management or a supervisor. The paucity of research on coworker perpetrated sexual harassment, along with the high prevalence of coworker perpetrated harassment, indicate a clear need to further investigate this form of workplace incivility.

Current Research

The purpose of the current research is to address several gaps in the literature. First, we aimed to test the casual effect of organizational policies on sexual harassment outcomes, extending the existing correlational literature. Second, we sought to use schema theory to test the interactive effects of organizational policy type and harassment severity on reporting outcomes, adding to what is known about the effects of organizational policies on harassment subtypes. Third, we intended to add to the literature on co-worker perpetrated sexual harassment, which is an understudied but more common form of sexual harassment than supervisor-perpetrated sexual harassment.

In the studies that follow, we tested the effects of various harassment policies (a zero-tolerance policy, a standard policy related to gender and harassment, or no policy) on bystander reporting of moderate sexual harassment (i.e., unwanted sexual attention that creates a hostile environment) and severe sexual harassment (i.e., sexual coercion implemented through a quid pro quo appeal). Because previous research has determined that sexual coercion is more likely to be perceived by both men and women participants and by the legal community as more severe and harassing than other types of sexual harassment (Fitzgerald and Ormerod 1991; Terpstra and Baker 1989), our first prediction was a main effect of harassment severity on bystander likelihood of reporting.

  • Hypothesis 1 Participants will be more likely to intend to report a severe incident of sexual harassment compared to a compared to a moderate instance of sexual harassment they ostensibly observed.

Based on research finding that zero-tolerance sexual harassment policies have the potential to increase workers’ sensitivity to incidents of harassment as well as bystander reporting of incidents of sexual harassment (e.g., Berdahl and Raver 2011; Buchanan et al. 2014), we also hypothesized the following main effect:

  • Hypothesis 2 Participants will be more likely to say they would report observed instances of sexual harassment when the company’s stated policy is explicitly zero-tolerance for sexual harassment, compared to when it is a standard policy or to when no policy is made known.

Our most important prediction, however, concerns how various company policies will affect bystander reporting of moderate compared to severe instances of sexual harassment. When social information is perceived as ambiguous, people look rely on schemas, scripts, and stereotypes to guide information processing and subsequent behaviors (Smith 1998). Evidence from correlational research supports the contention that zero-tolerance policies may increase reporting because they serve as a readily accessible schema to use for interpretation and decision-making. In particular, explicit zero-tolerance policies may increase the overall reporting rate of sexually harassing behaviors in organizations through an increased likelihood of labeling and reporting less severe, more ambiguous forms of hostile environment sexual harassment (e.g., non-sexual coercion) (Magley and Shupe 2005; Wilkerson 1999). Thus, in the tradition of social schemas, bystanders who witness moderate incidents of sexual harassment should be more likely to reference organizational policies in their interpretation, judgment, and decision making than those who witness severe incidents of sexual harassment. For these reasons, we predict the following interaction between harassment severity and company policy on likelihood of reporting an observed incident of sexual harassment:

  • Hypothesis 3. Policy type will have the strongest effect on bystander reporting likelihood for the moderate (vs. severe) harassment scenario. Specifically, in the severe harassment scenario, the policy will have a small but significant effect on likelihood of reporting, with the zero-tolerance policy being the most likely to result in reporting (H3a), and in the moderate harassment scenario, the company policy will have a stronger effect, again with the with the zero-tolerance policy being the most likely to result in participants saying they would report the incident (H3b).

Study 1

Our first experimental study sought to examine the effect of organizational policies and harassment severity on the likelihood of bystanders saying they would report the observed harassment to upper management in the organization.

Method

Sample

Our sample consisted of 219 undergraduate students enrolled in psychology courses at a large, R1, public research university located in the southeast of the United States. Of the participants in our sample, 55.25% reported being employed (n = 121) while 44.75% (n = 98) reported they were not employed when they completed the survey. The sample was about evenly split between men and women (55.25% women), and the mean age of the sample was 21.08 years (SD = 3.95). Additionally, 61.40% of the participants self-identified as Hispanic (n = 134), 17.5% were Caucasian (n = 38), 15.8% were African American (n = 34), and 5.3% were Asian (n = 12). The study was administered online using Qualtrics survey software, and informed consent was obtained from all individual participants included in the study. Participants were given course credit as an incentive for their participation.

Conditions

Men and women participants were assigned to 1 of 6 conditions in a 2(harassment severity: moderate vs. severe) × 3(type of organization sexual harassment policy: no policy, standard policy, zero-tolerance policy) fully between-subjects design.

Participants were exposed to the policy condition on the main webpage of a fictitious graphic design company. The webpage included a list of company services, a mission statement, and a policy section. All information about the fictitious company was consistent across the policy conditions except for the information contained in the policy section of the web page. Specifically, the policy statement included at the bottom of the company webpage for participants in the zero-tolerance policy condition was:

We have a zero-tolerance sexual harassment policy for all of our programs, which was developed by our service specialist and other staff and approved by the Executive Team. It outlines seven specific steps we will take to ensure gender equality in all programs and activities.

The standard policy condition statement that was included at the bottom of the webpage for those in the standard condition was:

This company is an equal opportunity organization and subscribes to Federal and State laws which forbid discrimination and harassment because of race, religion, color, sex, age, national origin, marital status, veteran status, and disability.

Lastly, the no policy condition statement included at the bottom of the webpage for participants who were in the no policy condition was:

We have a quality service policy that was developed by our Executive Team that ensures all of our employees are trained to provide the most effective and innovative work possible for our clients.

After reviewing the company web page, participants were randomly assigned to read either the severe or the moderate harassment scenario. The scenarios were pretested before the formal study to ensure that the severe scenario was seen as less ambiguous than the moderate scenario. Our pretest sample included 58 undergraduate students randomly assigned to read either the moderate or severe harassment scenario and rate the extent to which they perceived the behavior in the scenario to be sexual harassment on a 1(definitely not sexual harassment) to 7(definitely was sexual harassment) scale.Footnote 1 As expected, participants more often interpreted the behavior in the severe scenario as sexual harassment than the behavior in the moderate scenario (M = 6.09 vs. 4.46, SDs = 1.84 vs. 1.26), t(56) = 4.00, p < .001. The text for the final versions of the moderate and severe harassment conditions that were used in Study 1 was follows:

In an office setting, a female employee is sitting at her desk in her cubicle working on her computer with no one else visible in the office. A male coworker in her department who been working with her on a project for the past two months notices that she is alone, and then approaches her with a smile on his face and says “I just got to say, out of all of the female employees in the office, you have the best tits.” [Added in the severe scenario only: “I’m sure you realize that I have the deciding vote on whether or not you will be our next team leader. If you come to my place after work this Friday, wearing that red dress I like so much, my ballot will have your name on it.”] The female employee seems to become visibly nervous after the comment and nervously replies “thank you, I guess.” The male employee says “no problem” and then proceeds to talk about the group project that they are working on together. Assume you are an employee at the organization and you observed the entire interaction from your desk without either of the coworkers noticing you.

Participants were instructed to imagine they had anonymously observed this incident in the workplace. The female victim’s reaction was the same in both harassment scenarios with the only difference being type of behavior perpetrated. After participants finished reading their randomly assigned sexual harassment scenario, participants completed measures of the perceived seriousness of the incident and their likelihood of reporting the incident to upper management.

Measures

Likelihood of Reporting

Participants’ likelihood of reporting the incident of sexual harassment was measured using three items we developed drawing from past work on sexual harassment reporting (e.g., Gruber 1992; Lott et al. 1982). Specifically, likelihood of reporting was measured using the average score of the following three items, each measured on a 1(definitely would not report) to 7(definitely would report) scale: “how likely would you be to formally report the incident to your organization?,” “how likely would you be to formally report the incident to your supervisor?,” and “how likely would you be to formally report the incident to the human resources division in your organization?” Participants were instructed to answer with the assumption that no one else witnessed the behavior and that the employees in the scenario did not notice them. Cronbach’s alpha reliability estimate for the composite reporting measure was .93.

Manipulation Check

A modified version of the seriousness of offense scale (Gruber 1992) measured the extent to which participants perceived to behavior in the moderate vs. severe scenarios to be sexual harassment. It consisted of nine statements that asked participants to indicate the extent to which the behavior in the scenario was sexual harassment on a 1(strongly disagree) to 5(strongly agree) Likert-type scale. Sample scale items include “the male employee’s behavior is sexual harassment” and “the male employee’s behavior is offensive.” Cronbach’s alpha reliability estimate for the full 9-item seriousness of offense scale was .89. To ensure that respondents included in the sample accurately read the policy statement assigned to them on the organizational webpage, two policy manipulation check items were included at the end of each survey after reporting measures were completed and only the participants that passed the manipulation check items were included in the final sample. In these, participants were asked to select which content was included in the policy statement they read on the organizational webpage.

Demographic and Individual Difference Information

Demographic questions were also included in the survey, including questions on participant gender, race/ethnicity, age, employment, and level of education. We also assessed personal experience with sexual harassment as a potential control variable because those who have been targets of sexual harassment are more likely to label socio-sexual behaviors as harassing and are more likely to perceive such behaviors to be offensive (Blakely et al. 1995). To assess personal experience with sexual harassment, we used the16-item Sexual Experiences Questionnaire (SEQ) scale (Gettman and Gelfand 2007) which asks participants to recall the number of times they were subjected to various types of sexually harassing behavior (e.g., unwanted sexual attention, sexual coercion, etc.) at work in the past two years using a 1(never) to 5(daily) scale. In order to avoid a priming affect, the SEQ was administered to participants at the end of the survey, after all other measures had been completed. Cronbach’s alpha reliability estimate for the 16-item SEQ scale was .92.

Results

We conduced all analyses with SPSS 16. Results from an independent t-test analysis indicated that our manipulation of harassment severity was successful. Those who read the severe harassment scenario (M = 4.27, SD = .55) saw the offense in the scenario as more serious than those who read the moderate scenario (M = 3.46, SD = 1.03), t(219) = 3.71, p < .001), using the modified seriousness of offense scale as the dependent variable (Gruber 1992). To test H1, we conducted a two-way ANOVA with the likelihood of reporting composite measure as the dependent variable and harassment severity and participant gender as independent factors. Supporting H1, we found a main effect of sexual harassment severity on the likelihood of reporting composite measure, F(1, 218) = 39.72, R2 = .16, p < .001. Those who read the severe harassment scenario (M = 5.18, SD = 1.41) reported a higher likelihood of reporting the incident than those who read the moderate scenario (M = 3.87, SD = 1.66) (see Table 5). Next, we examined reporting likelihood by participant gender and found that the results held equally held for both women and men. Women who read the severe harassment scenario (M = 5.43, SD = 1.31) were more likely to indicate that they would report the incident than women who read the moderate harassment scenario (M = 4.08, SD = 1.57), t(146) = 5.11, p < .001. Men who read the severe harassment scenario (M = 4.88, SD = 1.48) were also more likely to indicate they would report the incident than the men who read the moderate harassment scenario (M = 3.60, SD = 1.75), t(146) = 3.94, p <. 001.

To test H2, we conducted a two-way ANOVA with likelihood of reporting as the outcome variable and policy type and participant gender as independent variables. Supporting H2, policy type predicted a significant amount of variance in likelihood of reporting sexual harassment, F(1, 217) = 13.33, R2 = .058, b = .34, p < .001. We then conducted post-hoc tests to examine how the likelihood of reporting varied across each of the three policy conditions. Overall, participants in the zero-tolerance policy condition were more likely to report the harassment (M = 5.24, SD = 1.48) than those in the no policy condition (M = 4.08, SD = 1.52), p < .001 (see Table 1). Similarly, participants in the zero-tolerance policy condition (M = 5.24, SD = 1.48) had a higher likelihood of reporting than those in the standard policy condition (M = 4.27, SD = 1.77), p < .001 (Table 1). However, there was not a significant difference in the overall likelihood of reporting between those in the standard policy condition and those in the no policy condition, p = .75 (see Table 1). Next, we examined the impact of policy type on reporting by participant gender. We found both women and men were more likely to report the sexual harassment in the zero-tolerance policy condition compared the standard policy condition or the no policy condition (see Fig. 1) (Table 2).

Table 1 Likelihood of reporting by organizational policy type in Study 1
Full size table
Fig. 1
figure 1

Mean ratings of likelihood to report sexual harassment by organizational policy, gender, and harassment severity in Study 1

Full size image
Table 2 Likelihood of reporting by organizational policy type in Study 2
Full size table

We next moved to test H3, that policy type would have the strongest effect on bystander reporting likelihood for moderate (vs. severe) harassment. To do so, we performed a hierarchical multiple regression with likelihood of reporting as the dependent variable, policy type, gender, and harassment severity on step 1, and the interaction between policy type and harassment severity on step 2 as the independent variables. As hypothesized, there was an interaction between policy type and harassment severity on likelihood of reporting. When entered on step 2, the interaction term predicted a significant amount of variance in reporting after controlling for policy type, previous experience of being sexually harassed, and harassment severity independently on step 1, F(3, 217) = 3.52, R2 = .226, ∆R2 = .018, b = .92, p = .031 (see Table 3).

Table 3 Summary of hierarchical regresssion analysis for variables predicinting likelihood of reporting sexual harassment in Study 1 (N = 219)
Full size table

As hypothesized, policy type had a stronger effect on the likelihood of reporting in the moderate harassment condition than in the severe harassment condition. For those who read the moderate harassment scenario, policy type predicted a significant amount of variance in reporting, F(1, 106) = 16.54, b = −.77, p < .001. Those in the zero-tolerance policy condition (M = 4.76, SD = 1.54) were more likely to report the incident of moderate harassment than those in the standard policy condition (M = 3.64, SD = 1.60), t(71) = 3.05, p = .003, and those in the no policy condition (M = 3.24, SD = 1.50), t(66) = 4.15, p < .001. Among those who read the severe harassment scenario, those in the zero-tolerance policy condition (M = 5.63, SD = 1.32) were more likely to report the incident than those in the standard policy condition (M = 4.13, SD = 1.29), t(74) = 3.54, p = .01. However, in the cause of witnessing severe harassment, the zero-tolerance policy condition did not appear to cause higher rates of expected reporting than the no policy condition, t(75) = 1.22, p = .23.

Finally, although rater characteristics are typically not as significant as behavioral characteristics in predicting perceptions of sexual harassment (Tata 2000), they may account for up to 10% of the variance in evaluations of sexual harassment (Gutek, 1995; Rotundo et al. 2001). For this reason, we also examined the effect of participant gender and experience with harassment on reporting outcomes, although they were not central to our predictions. There was a main effect of gender on reporting, F(1, 217) = 4.85, R2 = .022, p = .029. Consistent with previous studies indicating that women are more likely to perceive and report sexual harassment in the workplace (e.g., McCabe and Hardman 2005; Rotundo et al. 2001), women in our study were more likely to report sexual harassment than men across scenarios (M = 4.76, SD = 1.59 vs. M = 4.27, SD = 1.73), t(217) = 2.20, p = .03. However, we did not detect a significant interaction between policy, gender, and harassment severity, F(3215) = 1.68, p = .17. Previous experience with sexual harassment as measured by the 16-item SEQ scale did not significantly predict reporting of sexual harassment in Study 1, nor did it change the direction or the significance of any hypothesized analyses when included as a control variable.

Study 1 Discussion

Research has shown that organizational culture and sexual harassment severity are related to the pervasiveness of sexual harassment in a workplace, and may also play a role in how bystanders interpret and respond to harassment (Keyton et al. 2001). For example, observers of quid-pro-quo harassment are more likely to respond assertively than observers of hostile environment harassment (Benavides-Espinoza and Cunningham 2010), and organizational cultures that proactively promote gender equality greatly reduce sexual harassment in organizations (Timmerman and Bajema 2000). Our results shed new light on previous research by demonstrating that bystander perceptions of sexual harassment can be directly affected by organizational policies. Study 1 shows that formal zero-tolerance sexual harassment organizational policies, when salient, can increase the likelihood that less severe forms of sexual harassment (i.e., hostile environment harassment) are formally reported to organizations by employee bystanders. Our results also show that a salient zero-tolerance organizational policy can increase likelihood that more severe harassment (i.e., quid-pro-quo harassment) will be reported by bystanders compared to a standard policy statement, or the absence of a salient policy.

One significant limitation of the first study is that it was completed by a student sample rather than a sample of U.S. workers, potentially limiting the generalizability of the results to the workplace. Similarly, the three policy conditions in Study 1 were not based on actual policies of organizations where participants were employed. If participants had been given real policy statements from the organization in which they were employed, and were instructed to imagine the harassment scenario in their actual place of employment, concerns regarding the generalizability of the findings to the actual workforce would be reduced. Thus, it was important to see if the findings could be extended with a sample of employees using actual policy statements from their organization to create our policy conditions.

Study 2

To address the main limitation of Study 1, we conducted a second study with a sample of full-time employees at a single organization, and using actual excerpts from the organization’s employee manual to craft our policy conditions. Similar to Study 1, Study 2 sought to examine how the likelihood of reporting observations of coworker perpetrated male-to-female sexual harassment is influenced by harassment severity and company policy on sexual harassment.

Compared to Study 1, the design of Study 2 further reduced the likelihood that participants would be able to identify the research topic. Specifically, participants read about two other policies from the employee manual (i.e., an employee compensation policy and an employee training policy) along with the sexual harassment policy they were randomly assigned to. Participants in Study 2 were also given two additional coworker-related vignettes based on the two new policies they read (i.e., an employee compensation scenario and an employee training scenario) along with either the moderate or severe sexual harassment scenario from Study 1. Finally, to further decrease suspicions that the focal issue in our study was sexual harassment, all participants were falsely told they would be randomly assigned to answer questions on any one of the three coworker vignette scenarios they read. However, all participants in Study 2 only answered follow-up questions about the sexual harassment scenario.

Based on previous research, schema theories, and our Study 1 findings, we predicted the following for Study 2:

  • Hypothesis 4: When imagining having observed an incident of sexual harassment in their workplace, employee participants will be more likely to intend to report the severe (vs. the moderate) incident of sexual harassment to their organizations.

  • Hypothesis 5: Employees will be more likely to say they would report the observed sexual harassment when the policy excerpt from their organization reflects a zero-tolerance stance towards sexual harassment compared to when there is no policy excerpt, or when the policy excerpt on harassment is customary or standard.

  • Hypothesis 6. We predict a significant interaction between harassment severity and policy type on likelihood of reporting sexual harassment. Specifically, we predict policy type will have a stronger effect on reporting likelihood in the moderate harassment scenario than in the severe harassment scenario.

Method

Sample

Our sample for Study 2 consisted of 101 full-time male and female employees working in the Human Resource Department at a large public university located in the southeastern United States. Of the employees in the sample, 74.26% were female (n = 75) and 25.74% were male (n = 26). The average organizational tenure of employees was 6.93 years (SD = 6.52). Employee age ranged from 20 to 62 with a mean age of 36.62 years (SD = 3.95). Of the 99 participants who reported ethnicity, 58.4% were Hispanic (n = 59), 18.8% were Caucasian (n = 19), 12.9% were African American (n = 13), 3.0% were Asian (n = 3), and 5.0% were of another ethnicity (n = 5). Participants were recruited through emails sent to the Human Resource Department mailing consisting of 228 HR employees indicating a response rate of 44.30%. Participants completed the study anonymously online using Qualtrics survey software and informed consent was obtained from all individual participants included in the study. Participants were offered a $5 gift card as an incentive.

Conditions

Employee participants who agreed to complete the online survey emailed to them were randomly assigned to one of six conditions in a fully-between subjects 2(harassment severity: moderate vs. severe) × 3(type of organization sexual harassment policy: no policy, standard policy, zero-tolerance policy) study design. The same moderate and severe sexual harassment scenarios involving a female victim and a male coworker perpetrator from Study 1 were again used in Study 2. Unfortunately, the sample size included in Study 2, and the low proportion of men in the sample, did not permit us to analyze the role of participant gender. Therefore, the present findings may not generalize to samples of male employees. However, the gender composition of our sample in Study 2 was very similar to the national gender composition of human resource employees (Holland and Cortina 2016; Parkes and Davis 2013; ROCU 2014). Thus, while we were unable to examine the role of participant gender in this sample, the gender composition of our sample is high in external validity for this occupation.

The three policy conditions for Study 2 (zero-tolerance, standard, and no policy) were created using verbatim statements from the organization’s employee manual, and participants were told that the policy information they received was taken directly from their organization. Participants in the zero-tolerance policy condition read the following verbatim excerpt from the organization’s employee manual:

This organization has a zero-tolerance sexual harassment policy. To foster a climate that encourages prevention and reporting of sexual harassment, and related misconduct, the organization will actively promote zero-tolerance prevention efforts, respond to all reports promptly, provide interim protective measures to address safety and emotional well-being, and act in a manner that recognizes the inherent dignity of the individuals involved.

Participants in the standard policy condition read:

This organization prohibits all forms of discrimination and harassment based on an individual’s Protected Status, including, Age, Color, Creed, Disability, Gender, Gender Expression, Gender Identity, Genetic Information, National Origin, Race, Religion, Sex, Sexual Orientation, Veteran's Status and/or any other legally protected status.

Lastly, participants in the no policy condition read the following:

This organization seeks to provide quality, affordable services while maintaining four basic values we consider to be vital to our success. The conduct, ideals, and ethics that drive our operations are: quality, innovation, integrity, and service. These four values are core to this organization's mission of working in partnership with local organizations and institutions. Concerns regarding quality, innovation, integrity, and service are integrated into all aspects of this organization.

In addition to reading an excerpt of the organizational policy on sexual harassment, all participants were also given two other verbatim policy statements from the organization. One policy statement was related to employee development and the other was related to employee compensation. We included these additional policy statements to mitigate participant awareness of the study variables of interest.

After reviewing the organizational policies (the sexual harassment policy condition they were randomly assigned to and the two filler policies), participants read three vignettes. These vignettes were presented at random, and described workplace social interactions between coworkers. Participants were asked to imagine they were anonymous bystander observers in each of the three interactions. Two of the vignettes were filler vignettes related to employee development and compensation, again intended to disguise the true purpose of the study. The third vignette was either the moderate or severe sexual harassment scenario from Study 1. As in Study 1, participants were instructed to answer questions about all vignettes with the assumption that no one else witnessed the behavior and that the employees in the scenario did not notice them.

After reading the three policies and the three vignettes, all participants were informed that they would be randomly assigned to answer questions about one of the three scenarios. However, this was not true. Rather than being randomly assigned to answer questions about one of the three vignettes, all participants only received questions about the sexual harassment scenario. This was done to mitigate participants’ awareness that we were primarily interested in examining the topic of sexual harassment.

Measures

Likelihood of Reporting

Likelihood of reporting the incident of sexual harassment was measured using the same three items from Study 1. Cronbach’s alpha reliability estimate for the composite reporting scale was .96.

Manipulation Check

As in Study 1, we used a modified version of the seriousness of offense scale (Gruber 1992) to measure the extent to which participants perceived to behavior in the scenario to be sexual harassment. Sample scale items include “the male employee’s behavior is sexual harassment” and “the male employee’s behavior is offensive.” Cronbach’s alpha reliability estimate for the full 9-item seriousness of offense scale was .82.

Demographics and Individual Difference Information

At the end of the survey we obtained measures of participant gender, ethnicity, age, level of education, and the number of years and months participants worked at their current organization. As in Study 1, we also assessed personal experience with sexual harassment using the 16-item Sexual Experiences Questionnaire scale (Gettman and Gelfand 2007). Cronbach’s alpha reliability estimate for the full 16-item SEQ scale was .99.

Results

Results from an independent t-test analysis indicated that our manipulation of harassment severity was successful. Participants who received the severe harassment scenario saw the perpetrator’s behavior as more offensive using the seriousness of offense composite measure than those that read the moderate harassment scenario (Ms = 6.41 vs. 4.79, SDs = .79 vs. 1.78), t(99) = 5.87, p < .001. To test H4, we conducted a two-way ANOVA using harassment severity and policy type as the independent variables and the composite score of reporting as the dependent variable using SPSS 16. Since the majority of participants in this study were women, we did not have the opportunity to test for the effect of participant gender on reporting outcomes.

We found a significant main effect of harassment severity on reporting, F(1, 100) = 31.80, p < .001, R2 = .16. Those who read the severe harassment scenario (M = 5.60, SD = 1.12) were overall more likely to say they would formally report the harassment to their organization than those who read the moderate harassment scenario (M = 3.63, SD = 1.12), t(99) = 7.12, p < .01. Thus, the prediction that participants would generally be more likely to intend to report the severe (vs. the moderate) incident of sexual harassment (H4) was supported (see Fig. 2).

Fig. 2
figure 2

Mean ratings of likelihood of reporting sexual harassment by organizational policy and harassment severity in Study 2

Full size image

To test H5, that employees would be more likely to say they would report the sexual harassment when the policy excerpt from their organization reflects a zero-tolerance stance compared to the other two policies, we conducted a two-way factorial ANOVA using composite scores of likelihood of reporting as the criterion variable and policy type and harassment severity as the independent variables. We found a significant main effect of policy type on the likelihood of reporting, F(2, 100) = 7.15, p = .002 (see Table 6).

We then conducted post hoc analyses to examine how the likelihood of reporting varied across the three policy conditions. Across the moderate and severe harassment scenarios, those in the zero-tolerance condition (M = 5.27, SD = 1.16) were significantly more likely to report the behavior they read about to their organizations than those in the standard policy condition (M = 4.28, SD = 1.94), t(62) = 2.47, p = .017. Similarly, those in the zero-tolerance condition were significantly more likely to report than those in the no policy condition (M = 4.26, SD = 1.73) across the moderate and the severe harassment scenarios, t(67) = 2.81, p = .005. Lastly, there was not a significant difference in reporting between those in the standard group and those in the no policy group across the moderate and the severe harassment scenarios, t(67) = .05, p = .964. Since there was a significant main effect of policy type on likelihood of reporting across both levels of harassment, H5 was supported (see Table 2).

H6 specified that we would detect a significant interaction between harassment severity and policy type on likelihood of reporting sexual harassment, with policy type having a stronger effect on reporting likelihood in the moderate (vs. severe) harassment scenario. To test this, we conducted a two-way factorial ANOVA using harassment severity and policy type as the independent variables and the composite score of reporting as the dependent variable. We found a significant interaction between policy type and harassment severity on likelihood of reporting after controlling for policy type and harassment severity independently, F(2, 100) = 6.32, p = .003. Subsequently, a hierarchical linear regression analysis with harassment severity, previous experience being sexually harassed, and policy type entered on step 1 and the interaction term entered on step 2 indicated that the interaction accounted for a significant amount of unique incremental variance of likelihood to report after controlling for the type of harassment, previous experience of being sexually harassed, and policy type independently, F(3, 100) = 6.32, ∆R 2 = .046, b = .89, p = .006 (see Table 4).

Table 4 Summary of hierarchical regression analysis for variables predicting likelihood of reporting sexual harassment in Study 2 (N = 101)
Full size table
Table 5 Study 1 two-way ANOVA results for likelihood of reporting sexual harassment by organizational policy type and sexual harassment severity
Full size table
Table 6 Study 2 two-way ANOVA results for likelihood of reporting sexual harassment by organizational policy type and sexual harassment severity
Full size table

We then ran two subsequent ANOVAs, one for those who received the moderate harassment scenario and one for those who received the severe harassment scenario. For those who received the moderate harassment scenario, results indicated a significant main effect of policy type on reporting, F(2, 51) = 12.13, p < .001, partial eta squared = .33. More specifically, those in the zero-tolerance condition (M = 4.94, SD = 1.29) were significantly more likely to report moderate harassment than those in the standard policy condition (M = 2.87, SD = 1.48), t(30) = 4.24, p < .001, or those in the no policy condition (M = 3.10, SD = 1.27), t(35) = 4.36, p < .001 (see Fig. 2).

Contrary to the effects of policy on the reporting of moderate harassment, we did not find an effect of policy type on reporting likelihood for severe harassment, F(2, 49) = .03, p = .97, partial eta squared = .001. Those in the zero-tolerance policy condition (M = 5.64, SD = 0.90) were not significantly more likely to report severe sexual harassment than those in the standard policy condition (M = 5.54, SD = 1.39), t(30) = .23, p = .82, or those in the no policy condition (M = 5.63, SD = 1.05), t(30) = .05, p = .96. Thus, supporting H6, the effect of the policy on reporting was significant and much stronger in the moderate harassment scenario than it was in the severe harassment scenario (Tables 5 and 6).

Previous experiences of sexual harassment did not predict a significant amount of unique variance in reporting likelihood, F(1, 95), = 1.05, R2 = .011, p = .31. Previous experience of sexual harassment also did not affect the significance of the main effect of severity on reporting, the main effect of policy type on reporting, or the interaction between policy and severity on reporting when it was included as a control variable. At the end of the survey, we also asked the participants in the sample how familiar they were with the policies in their organization’s employee manual as well as how familiar they were with the sexual harassment policies at their organization on a scale from 1 (not at all familiar to 7(extremely familiar). Results indicated that the employees in the sample were only somewhat familiar with the policies in their organization’s employee manual (M = 5.05, SD = 1.20) and the sexual harassment policies at their organization (M = 4.85, SD = 1.47).

Study 2 Discussion

As expected, and similar to Study 1, our second study that employees who read the severe harassment scenario involving coworker perpetrated quid-quo harassment indicated a higher likelihood of reporting the behavior to their organizations than those who read the moderate harassment scenario involving coworker perpetrated hostile environment harassment. Similar to the first study, there also was a main effect of policy type on the likelihood of reporting in Study 2. Employee participants in the second study said they would be more likely to report sexual harassment when they read that their organization has a zero-tolerance policy (vs. a standard policy or no explicit policy). Finally, as predicted based on schema theory, the zero-tolerance policy was particularly effective in increasing reporting likelihood among employees who received the moderate harassment scenario compared to the severe harassment scenario.

However, contrary to our findings from Study 1, we did not find an effect of policy type on likelihood of reporting the severe harassment incident. Although we still expected to find a difference in reporting between those in the zero-tolerance condition and those in the other two policy conditions for the severe harassment, these results affirm our suspicion that most employees are able to effectively identify overt quid-pro-quo type harassment as inappropriate, and are likely to report it without needing to reference their organization’s policy to guide their decision. Additionally, it is possible these contrary findings may be due to the fact that the Study 2 participants were human resources employees. Regardless of the cognitive salience of organizational polices, human resources employees have a generally clear understanding of severe types of sexual harassment, and the importance of formally reporting it to their organization. Overall, the findings from Study 2 again highlight the potential importance that zero-tolerance policies could play in the formal reporting of subtler forms of sexual harassment in organizations.

Since moderate forms of harassment occur much more frequently in organizations and are more likely to go unreported compared to more overt, severe types of sexual harassment (e.g., quid pro quo) implementing explicit zero-tolerance policy statements should help substantially increase the likelihood that the most common forms of sexual harassment are formally reported brought to the attention of organizations which. In time, this may discourage future harassing behavior from occurring by lowering employees’ perceptions of organization’s tolerance for sexual harassment.

General Discussion

The current study addresses important gaps in the sexual harassment reporting and organizational policy literature by examining bystander perceptions and likelihood of reporting coworker perpetrated sexual harassment. Coworker perpetrated harassment remains one of the most pervasive forms of sexual harassment that is often unreported by targets (Berdahl and Raver 2011; Firestone et al. 2012). Additionally, the current study gives novel insights regarding how organizational sexual harassment policies may impact bystander reporting of coworker perpetrated sexual harassment with varying levels of severity or ambiguity.

The results of the current studies are consistent with previous research suggesting that organizations can influence the pervasiveness and outcomes of sexual harassment through the implementation of formal policies. The type of policies held in organizations have been related to the prevalence of harassment, the likelihood of reporting incidents of harassment, how sexual harassment is interpreted by the target, and how the target responds to the behavior (Gruber 1998; Miner-Rubino and Cortina 2004). The present studies provide evidence that policies can cause changes in bystander likelihood of reporting. Our results also demonstrate that zero-tolerance policies may be particularly effective in increasing the reporting of moderate or subtle forms of sexual harassment, consistent with knowledge about the influence of schemas in interpretations of and responses to ambiguous situations.

Previous research has found that the likelihood of reporting harassment is influenced by the perceived level of risk about making a complaint, the likelihood that complaints are taken seriously, and the likelihood the perpetrators will be punished. Zero-tolerance organizational policies may also increase bystander and personal report of sexual harassment because they are perceived by employees to indicate a low risk for reporting a complaint, that complaints will be taken seriously at the organization, and an increased likelihood that the perpetrator will actually be punished (Hulin et al. 1996; McLaughlin et al. 2012). Thus, our results seem to support the contention that implementing explicit zero-tolerance organizational policies should increase reporting behavior in organizations which, in turn, may reduce the prevalence of future sexual harassment behaviors and the detrimental outcomes associated with them (Berdahl and Raver 2011).

Implications for Practice

Many organizations have a zero-tolerance sexual harassment policy that may include mandated reporting or training. However, it is the salience of such organizational policies, along with norms of policy adherence, that may ultimately determine the extent that employees exhibit sexually harassing behavior and the likelihood that such behavior is reported or challenged. Both policy salience and norms of policy adherence stress that the sexual harassment policy is an organizational priority, and one that is respected by employees (McLaughlin et al. 2012). Our findings suggest that even if organizations are legally required to implement and adhere to reporting policies and procedures, workplaces may not be conducive to reporting sexual harassment unless the organizational policies are perceived to exhibit low tolerance of sexual harassment (McLaughlin et al. 2012).

In recent years, organizations have taken a greater interest in promoting and prioritizing employee “whistleblowing” (i.e., the disclosure of wrongdoing) related to sexually harassing behaviors (Buchanan et al. 2014; Keenan 2002). This includes designating departments, individuals, or parties within the organization for the receipt of whistleblowing reports and documentation, which are expected to and capable of taking action to correct the wrongdoings (Berdahl and Raver 2011; Buchanan et al. 2014; Firestone et al. 2012). To understand the contextual factors that impact sexual harassment whistleblowing and the whistleblowing of other counterproductive workplace behaviors, research has examined organizational structures, industry type, and culture (Buchanan et al. 2014; Keenan 2002, King 1999; Mesmer-Magnus and Viswesvaran, 2005;Miceli and Near 1992; Sims and Keenan 1999) in both the public sector (e.g., Miceli et al. 1999) and the private sector (e.g., Keenan and McLain 1992). Thus, another contribution of the current studies to the sexual harassment reporting research is that they fill a gap on how organizational policies may impact bystander reporting behavior for both moderate and severe sexually harassing behaviors. This can help guide the development and implementation of effective sexual harassment reporting policies.

Limitations and Future Research Directions

Although the results of these two studies help address crucial gaps in the existing sexual harassment literature, one of the major limitations of the current studies is that the harassment conditions were manipulated using vignettes. Participants reporting intentions using vignettes may be different than if the behaviors described in the scenarios were actually witnessed in an organizational setting. Future research is needed to clarify this relationship by actually displaying the perpetrated harassment behavior to participant observers either in real life and assessing actual reporting behaviors. Future studies could also seek to examine how the individual characteristics of the perpetrator and target involved in coworker harassment scenarios (e.g., the gender of the perpetrator) influence bystander reporting behavior and perceived seriousness of different types of harassment. Also, although versions of hostile environment sexual harassment and quid pro quo sexual harassment were included in the current study, it is also possible that other variations on these types, and other distinct types of sexual harassment, may be differently impacted by organizational policy types.

Given that sexual harassment policies are generally written for the target of the harassment, and how she/he can or should report incidents of sexual harassment, it is possible that companies seeking to increase and encourage bystander reporting of sexual harassment may do so by adjusting their policies to explicitly state that all bystander employees who observe sexual harassment should report it (Berdahl and Raver 2011; Buchanan et al. 2014). However, since the content of the zero-tolerance sexual harassment organizational policies in our studies did not explicitly indicate that bystander employees should formally report it to their organization, future research on zero-tolerance policies that do and do not include clear reporting guidelines for bystanders is needed. Additionally, to our knowledge, previous research on sexual harassment reporting has not used hypothetical scenarios to examine the likelihood that observers would approach the victim to encourage them to report the behavior. As a result, future work on how organizational policy may impact the likelihood of bystanders to encourage targets of harassment to formally report the behavior could expand the current study’s findings.

One of the main limitations of Study 2 is a potential lack of generalizability due to using a sample of human resources professionals. Human resources professionals may have a better understanding of sexual harassment policies and sexual harassment behaviors than employees in other divisions of an organization (Bernadette van Rijn et al. 2013; Parkes and Davis 2013; Van Den Ouweland and Van den Bossche 2017). The tendency for human resource employees to have a greater understanding of sexual harassment organizational policies and behaviors may due to occupation-specific learning incentives and career motivation (Bernadette van Rijn et al. 2013; Van Den Ouweland and Van den Bossche 2017). They may also be more vigilant in their application of the policies, as they have potentially fuller understanding of their scope and consequences (Parkes and Davis 2013). However, despite the fact that we used a sample of HR professionals, we still found that organizational policy significantly impacted expected bystander reporting, and did so differentially for moderate versus severe forms of harassment. This speaks to the importance of both organizational policies and harassment severity in affecting the bystander reporting of sexual harassment; we expect the effects of these variables on bystander reporting may be even more profound among those who are not HR professionals, or who are not as familiar with and attentive to organizational policies. To further asses the generalizability of the findings of the second study, future studies should examine how the salience of organizational sexual harassment policies influences bystander reporting of harassment in samples of employees from a variety of occupations and industries.

Another important limitation of Study 2 was that our sample mainly consisted of women respondents, and was in a stereotypically-feminine work domain (human resources). Therefore, our sample may not generalize to samples of mainly men, or to traditionally masculine domains where sexual harassment is particularly prevalent. For example, 69% of women in military samples reported experiencing sexually harassing behavior compared to 58% of women in the academic sector, 46% in the private sector, and 43% in the government sector (Ilies et al. 2003). The increased prevalence of harassment in male-dominated domains and among agentic women who violate feminine gender norms may signify male resentment and a form of retaliation to competent women succeeding in traditionally masculine settings (Leskinen et al. 2011; Eagly and Carli 2007; Carli 2004). Thus, future studies should seek to examine how the salience of sexual harassment policies influences bystander reporting behavior of moderate and severe forms of harassment in highly masculine domains.

Conclusion

The present study contributes to the body of sexual harassment research by examining bystander likelihood of reporting of coworker perpetrated sexual harassment, which remains one of the most pervasive and underreported forms of sexual harassment (Firestone et al. 2012; Berdahl and Raver 2011). Additionally, the current study contributes to the sexual harassment literature by examining how organizational sexual harassment policies, including both standard/generic policies and zero-tolerance policies, affect the likelihood of bystanders reporting coworker perpetrated “quid pro quo” and “hostile environment” sexual harassment. Lastly, this work presents novel findings on how the salience of explicit zero-tolerance organizational sexual harassment policies may impact the likelihood that bystanders report moderate and severe forms of sexual harassment to their organization.