The following commentary was inadvertently omitted from Volume 12, Issue 1, as a response to the focal article https://doi.org/10.1017/iop.2018.155, “#Ustoo: How I-O Psychologists Can Extend the Conversation on Sexual Harassment and Sexual Assault Through Workplace Training.” SIOP regrets the error.
The focal article by Medeiros and Griffith (Reference Medeiros and Griffith2019) discusses some ways to reinforce longitudinal benefits and prolong behavioral change of sexual harassment training. Although we generally agree with the suggestions, we noticed that the focal article did not address two of the most important elements to understanding the workplace sexual harassment: power and dependence. Weber (Reference Weber1978) defined power as “the probability that one actor within a social relationship will be in a position to carry out his will despite resistance (p. 53).” Research has shown that a majority of sexual harassment incidents occur when perpetrators with higher power coerce sexual behaviors onto victims with lower power (e.g., Thacker & Ferris, Reference Thacker and Ferris1991). For example, it is more common for supervisors than their subordinates to be a perpetrator rather than a victim. Likewise, it is more common for teachers than their students to be a perpetrator rather than a victim. One common characteristic of the two examples is that those who are more likely to be a perpetrator can exert power over their weaker counterparts.
Most of the sexual harassment trainings have focused on informing the employees with the definitions of harassment, severe consequences of sexual harassment, and administrative steps to report such incidents (Berdahl & Raver, Reference Berdahl, Raver, Zedeck and Zedeck2011). Despite their practical usefulness, we argue that such trainings are unlikely to make a significant change. In general, the prevalence of workplace harassment does not originate from employees’ lack of knowledge in identifying the sexually harassing behaviors. Rather, it results from a lack of environmental resources that can be utilized to effectively punish sexual harassers and discourage victim tolerance (Roehling & Huang, Reference Roehling and Huang2018). Lack of such resources is closely associated with the power distance and dependence between a harasser and a victim at workplace. The discussion of these critical elements in understanding sexual harassment at the workplace, however, was largely overlooked in the focal article. Thus, in this commentary, we emphasize the need for understanding power and dependence to effectively reduce workplace sexual harassment.
Differential power structure in organizations
Research has indicated that there are some industrial and organizational characteristics that contribute to the prevalence of sexual harassment. Many of them concern differential power structure in organizations (Berdahl & Raver, Reference Berdahl, Raver, Zedeck and Zedeck2011). For instance, industries with higher status or power hierarchy have a higher rate of reported sexual harassment than their counterparts (Ilies, Hauserman, Schwochau, & Stibal, Reference Ilies, Hauserman, Schwochau and Stibal2003). A meta-analysis based on more than 86,000 respondents from four different industries suggests that military samples (69%) reported the highest number of sexual harassment incidents, followed by academic samples (58%; Ilies et al., Reference Ilies, Hauserman, Schwochau and Stibal2003). Typically, there are well-established power structures and hierarchies in both military and academia. In addition, service-oriented industries, such as hotels and restaurants, were also found to have high level of reported sexual harassments (Yagil, Reference Yagil2008). Servers in these industries, usually seen as being relatively lower on power, were found to face sexual harassment or mistreatment enacted by customers on a daily basis (e.g., Barling, Rogers, & Kelloway, Reference Barling, Rogers and Kelloway2001; Korczynski & Evans, Reference Korczynski and Evans2013). Research found that an identical sexual harassing behavior was perceived less negatively if it was done by a customer than by a service employee because of the perception that customers are entitled to exert power (Madera, Podratz, King, & Hebl, Reference Madera, Podratz, King and Hebl2007; Poulston, Reference Poulston2008).
A related factor, job gender context, which refers to the sex ratio of one’s work group and the extent to which one’s occupation is male dominant, was found to be related to the prevalence of sexual harassment (Johnson, Kirk, & Keplinger, Reference Johnson, Kirk and Keplinger2016; Willness, Steel, & Lee, Reference Willness, Steel and Lee2007). Berdahl (Reference Berdahl2007) found that more women from male-dominating organizations reported having experienced sexual harassment than those from female-dominating organizations. Roth (Reference Roth2007) also suggested that organizations with a “macho” culture tend to normalize sexual harassment, contributing to a high prevalence of sexual harassment. Empirical and meta-analytic findings have confirmed that job gender context is an important risk factor for sexual harassment (Fitzgerald, Drasgow, Hulin, Gelfand, & Magley, Reference Fitzgerald, Drasgow, Hulin, Gelfand and Magley1997; Willness et al., Reference Willness, Steel and Lee2007).
The top management team, which usually consists of the most powerful figures in an organization, plays a critical role in workplace sexual harassment. Organizational climate for sexual harassment refers to a perceived organizational-level tolerance of sexual harassment at the workplace (Fitzgerald et al., Reference Fitzgerald, Drasgow, Hulin, Gelfand and Magley1997). This climate may contain perceived risks for the target’s complaints, lack of sanctions against perpetrators, and perceptions that one’s complaint will not be taken seriously (Hulin, Fitzgerald, & Drasgow, Reference Hulin, Fitzgerald, Drasgow and Stockdale1996). Hulin et al. (Reference Hulin, Fitzgerald, Drasgow and Stockdale1996) found that sexual harassment is more common if an organization has a top management team that tends to tolerate it. Both empirical and meta-analytic findings have confirmed that organizations with a climate for tolerating sexual harassment are more likely to have a high prevalence of workplace sexual harassment (Fitzgerald et al., Reference Fitzgerald, Drasgow, Hulin, Gelfand and Magley1997; Fitzgerald, Gelfand, & Drasgow, Reference Fitzgerald, Gelfand and Drasgow1995; Willness et al., Reference Willness, Steel and Lee2007).
In addition, sexual harassment could be more common in organizations with high levels of interpersonal competition, because employees in highly competitive workplaces tend to be more protective of one’s status than those in less competitive workplaces (Berdahl & Raver, Reference Berdahl, Raver, Zedeck and Zedeck2011). Moreover, Bowling and Beehr (Reference Bowling and Beehr2006) argued that workplaces characterized by high-level stress tend to predict more frequent sexual harassment incidents. Supporting the hypothesis, their meta-analytic results showed that sexual harassment was positively predicted by job stressors such as role conflict, role ambiguity, role overload, and work constraints, and was negatively predicted by autonomy (Bowling & Beehr, Reference Bowling and Beehr2006).
Theoretical explanations of why victims stay quiet
Multiple perspectives can provide insights to understand why people with less power may tolerate and remain silent while undergoing misconduct enacted by those with more power. A greater power distance between target and perpetrator may pressure the target into staying silent. Such a power distance can be conceptualized onto various dimensions, including gender, race, age, and social class. Previous studies have shown that vulnerable or low-skilled individuals are more likely to become targets of sexual harassment because of their lack of power (Uggen & Blackstone, Reference Uggen and Blackstone2004). Evidence has also indicated that younger women and minority racial groups are at a higher risk of being sexually harassed than older women and the racial majority. As discussed earlier, in service industry, customers are usually perceived to have higher power, thereby making mistreatments enacted by customers more tolerable than those by service employees (Poulston, Reference Poulston2008). Hence, the power distance between the two parties within a particular dimension may create an environment for the target to stay silent when undergoing mistreatments including sexual harassment.
Resource dependency between the two parties is another critical factor. According to Emerson (Reference Emerson1962), social power can create asymmetric control over resources in social relations. This asymmetry creates unequal degree of dependency between the two parties; the lower power parties may heavily depend on the higher power parties for valuable resources, while the higher power parties do not depend on the lower power parties (Magee & Galinsky, Reference Magee and Galinsky2008). For example, subordinates rely on supervisors for promotion opportunities; servers depend on customers to earn tips. On the other hand, higher power parties rarely depend on lower power parties to gain monetary resources. In addition, higher power parties tend to display a greater interpersonal variability in behaviors than the lower power parties, implying that their behaviors are less constrained by the environment. Hollander’s theory of idiosyncratic credits (Reference Hollander1958) suggests that people with higher power tend to be less confined by the social norm, thereby behaving as they please. Therefore, it appears that the more a target depends on a perpetrator for resources, the higher likelihood that misbehaviors by the higher power parties will be tolerated or accepted.
Furthermore, a target’s silence may be mistakenly taken as a sign of appropriateness for the misbehavior, which encourages the perpetrator to continue his or her behavior. Bargh and Raymond (Reference Bargh and Raymond1995) suggest that some perpetrators are unaware of the inappropriateness of their actions.
Practical recommendations
Following our argument that power and dependency are two critical aspects in understanding workplace sexual harassment, we make practical recommendations for designing better sexual harassment training in this section.
Clearly define the power boundary of higher power parties
In sexual harassment training, it is not only important to educate employees with clear definitions of inappropriate behaviors but also to inform them about the legitimate boundaries of authority held by those with greater power. This may begin by reinforming employees of their job descriptions and training them on the issues of power abuse with detailed examples. The purpose of doing so is to prevent potential victims, most likely the lower power employees, from tolerating or licensing misconduct by those with greater power. Although the people with high power have more legitimate authority to influence those with low power to a certain degree, it is essential for both parties to keep in mind the boundaries of their influence and understand that it is not their obligation to comply with commands beyond their job duties. Thus, including this aspect into sexual harassment training would potentially improve its effectiveness.
Loosen the dependency of lower power parties on higher power parties
Emerson (Reference Emerson1962) argues that a power dynamic represents the relative state of dependency between two parties. The extent to which A has power over B is largely determined by the extent to which B depends on A for certain resources and the availability of alternative suppliers of the same resource to which B can have access (Emerson, Reference Emerson1962). If B is in need of a valuable resource and can only obtain it from A, meaning this resource is important and there is a lack of alternative supply, we can say that B is highly dependent on A. In the organizational context, if an employee’s chance of career advancement is important to him/her, and is only determined by a direct supervisor’s evaluation, the employee may attempt to manage positive impressions and show a high compliance to the direct supervisor. The more dependent the employee’s career-advancing resources are on the supervisor, the more power the supervisor has over the employee. This further increases the likelihood of the supervisor exploiting the employee without facing much resistance.
Expanding availability of the resources given to those with lower power can potentially disconnect the power structure (Emerson, Reference Emerson1962). From the organizational and managerial perspective, it is essential to ensure that one’s career success is not fully contingent on decisions made by one, single, powerful person. We argue that it is important to assign multiple mentors to look after the lower-level employees and assign multiple raters to evaluate employees’ performances with standardized criteria. A multiple mentor system is not only beneficial to the fairness of performance appraisal but also essential to improving the effectiveness of sexual harassment training by intervening in the power and dependency structure in workplace.
Increase accesses for reporting incidents
Outside of the training context, it is also important to ensure that there is easy access to methods for the targets and bystanders to report sexual harassment incidents. The reports should be filed through easy formal procedures, such as filling out an online form to immediately report sexual harassment to the human resources department. If the report has to go through direct supervisors, the reporting system may not be as helpful, especially when a direct supervisor is the perpetrator.
Although it may be helpful to provide formal procedures for reporting sexual harassment incidents, individuals with lower power may be reluctant to utilize them if they fear retaliation or being ignored afterward. Therefore, perceived availability of informal procedures outside of the training context can be equally important in addressing the sexual harassment problem. Informal resources should include coaching, counseling, resolving harassment situations, and instrumental or emotional support by supervisors and coworkers (Berdahl & Raver, Reference Berdahl, Raver, Zedeck and Zedeck2011).
Fair performance appraisal and compensation
It is also critical that organizations ensure equal compensation to employees with the same job positions, regardless of their demographic backgrounds. In order to ensure such gender pay equality, organizations should accurately represent each employee’s merit in the compensation structure. Organizations should also ensure that raters of performance appraisal are not biased toward any demographic group, as both unfavorable and favorable biases may jeopardize the fair ratings (Harris, Reference Harris1994). When designing criteria for performance appraisal, organizations should ensure that the criteria are relevant to the job and are unbiased toward any group regardless of the level of power. Moreover, organizations are encouraged to provide a transparent compensation and promotion system to enhance the perception of fairness among their employees.
Conclusion
In conclusion, we argue that organizations could minimize the occurrence of sexual harassment and increase the effectiveness of sexual harassment training by understanding the power and dependence dynamic reflected by organizational structures. Moreover, ensuring that power relations are well-controlled in all environments outside training can facilitate enduring transfer effects from the training, further reducing sexual harassment in the long term.
The following commentary was inadvertently omitted from Volume 12, Issue 1, as a response to the focal article https://doi.org/10.1017/iop.2018.155, “#Ustoo: How I-O Psychologists Can Extend the Conversation on Sexual Harassment and Sexual Assault Through Workplace Training.” SIOP regrets the error.
The focal article by Medeiros and Griffith (Reference Medeiros and Griffith2019) discusses some ways to reinforce longitudinal benefits and prolong behavioral change of sexual harassment training. Although we generally agree with the suggestions, we noticed that the focal article did not address two of the most important elements to understanding the workplace sexual harassment: power and dependence. Weber (Reference Weber1978) defined power as “the probability that one actor within a social relationship will be in a position to carry out his will despite resistance (p. 53).” Research has shown that a majority of sexual harassment incidents occur when perpetrators with higher power coerce sexual behaviors onto victims with lower power (e.g., Thacker & Ferris, Reference Thacker and Ferris1991). For example, it is more common for supervisors than their subordinates to be a perpetrator rather than a victim. Likewise, it is more common for teachers than their students to be a perpetrator rather than a victim. One common characteristic of the two examples is that those who are more likely to be a perpetrator can exert power over their weaker counterparts.
Most of the sexual harassment trainings have focused on informing the employees with the definitions of harassment, severe consequences of sexual harassment, and administrative steps to report such incidents (Berdahl & Raver, Reference Berdahl, Raver, Zedeck and Zedeck2011). Despite their practical usefulness, we argue that such trainings are unlikely to make a significant change. In general, the prevalence of workplace harassment does not originate from employees’ lack of knowledge in identifying the sexually harassing behaviors. Rather, it results from a lack of environmental resources that can be utilized to effectively punish sexual harassers and discourage victim tolerance (Roehling & Huang, Reference Roehling and Huang2018). Lack of such resources is closely associated with the power distance and dependence between a harasser and a victim at workplace. The discussion of these critical elements in understanding sexual harassment at the workplace, however, was largely overlooked in the focal article. Thus, in this commentary, we emphasize the need for understanding power and dependence to effectively reduce workplace sexual harassment.
Differential power structure in organizations
Research has indicated that there are some industrial and organizational characteristics that contribute to the prevalence of sexual harassment. Many of them concern differential power structure in organizations (Berdahl & Raver, Reference Berdahl, Raver, Zedeck and Zedeck2011). For instance, industries with higher status or power hierarchy have a higher rate of reported sexual harassment than their counterparts (Ilies, Hauserman, Schwochau, & Stibal, Reference Ilies, Hauserman, Schwochau and Stibal2003). A meta-analysis based on more than 86,000 respondents from four different industries suggests that military samples (69%) reported the highest number of sexual harassment incidents, followed by academic samples (58%; Ilies et al., Reference Ilies, Hauserman, Schwochau and Stibal2003). Typically, there are well-established power structures and hierarchies in both military and academia. In addition, service-oriented industries, such as hotels and restaurants, were also found to have high level of reported sexual harassments (Yagil, Reference Yagil2008). Servers in these industries, usually seen as being relatively lower on power, were found to face sexual harassment or mistreatment enacted by customers on a daily basis (e.g., Barling, Rogers, & Kelloway, Reference Barling, Rogers and Kelloway2001; Korczynski & Evans, Reference Korczynski and Evans2013). Research found that an identical sexual harassing behavior was perceived less negatively if it was done by a customer than by a service employee because of the perception that customers are entitled to exert power (Madera, Podratz, King, & Hebl, Reference Madera, Podratz, King and Hebl2007; Poulston, Reference Poulston2008).
A related factor, job gender context, which refers to the sex ratio of one’s work group and the extent to which one’s occupation is male dominant, was found to be related to the prevalence of sexual harassment (Johnson, Kirk, & Keplinger, Reference Johnson, Kirk and Keplinger2016; Willness, Steel, & Lee, Reference Willness, Steel and Lee2007). Berdahl (Reference Berdahl2007) found that more women from male-dominating organizations reported having experienced sexual harassment than those from female-dominating organizations. Roth (Reference Roth2007) also suggested that organizations with a “macho” culture tend to normalize sexual harassment, contributing to a high prevalence of sexual harassment. Empirical and meta-analytic findings have confirmed that job gender context is an important risk factor for sexual harassment (Fitzgerald, Drasgow, Hulin, Gelfand, & Magley, Reference Fitzgerald, Drasgow, Hulin, Gelfand and Magley1997; Willness et al., Reference Willness, Steel and Lee2007).
The top management team, which usually consists of the most powerful figures in an organization, plays a critical role in workplace sexual harassment. Organizational climate for sexual harassment refers to a perceived organizational-level tolerance of sexual harassment at the workplace (Fitzgerald et al., Reference Fitzgerald, Drasgow, Hulin, Gelfand and Magley1997). This climate may contain perceived risks for the target’s complaints, lack of sanctions against perpetrators, and perceptions that one’s complaint will not be taken seriously (Hulin, Fitzgerald, & Drasgow, Reference Hulin, Fitzgerald, Drasgow and Stockdale1996). Hulin et al. (Reference Hulin, Fitzgerald, Drasgow and Stockdale1996) found that sexual harassment is more common if an organization has a top management team that tends to tolerate it. Both empirical and meta-analytic findings have confirmed that organizations with a climate for tolerating sexual harassment are more likely to have a high prevalence of workplace sexual harassment (Fitzgerald et al., Reference Fitzgerald, Drasgow, Hulin, Gelfand and Magley1997; Fitzgerald, Gelfand, & Drasgow, Reference Fitzgerald, Gelfand and Drasgow1995; Willness et al., Reference Willness, Steel and Lee2007).
In addition, sexual harassment could be more common in organizations with high levels of interpersonal competition, because employees in highly competitive workplaces tend to be more protective of one’s status than those in less competitive workplaces (Berdahl & Raver, Reference Berdahl, Raver, Zedeck and Zedeck2011). Moreover, Bowling and Beehr (Reference Bowling and Beehr2006) argued that workplaces characterized by high-level stress tend to predict more frequent sexual harassment incidents. Supporting the hypothesis, their meta-analytic results showed that sexual harassment was positively predicted by job stressors such as role conflict, role ambiguity, role overload, and work constraints, and was negatively predicted by autonomy (Bowling & Beehr, Reference Bowling and Beehr2006).
Theoretical explanations of why victims stay quiet
Multiple perspectives can provide insights to understand why people with less power may tolerate and remain silent while undergoing misconduct enacted by those with more power. A greater power distance between target and perpetrator may pressure the target into staying silent. Such a power distance can be conceptualized onto various dimensions, including gender, race, age, and social class. Previous studies have shown that vulnerable or low-skilled individuals are more likely to become targets of sexual harassment because of their lack of power (Uggen & Blackstone, Reference Uggen and Blackstone2004). Evidence has also indicated that younger women and minority racial groups are at a higher risk of being sexually harassed than older women and the racial majority. As discussed earlier, in service industry, customers are usually perceived to have higher power, thereby making mistreatments enacted by customers more tolerable than those by service employees (Poulston, Reference Poulston2008). Hence, the power distance between the two parties within a particular dimension may create an environment for the target to stay silent when undergoing mistreatments including sexual harassment.
Resource dependency between the two parties is another critical factor. According to Emerson (Reference Emerson1962), social power can create asymmetric control over resources in social relations. This asymmetry creates unequal degree of dependency between the two parties; the lower power parties may heavily depend on the higher power parties for valuable resources, while the higher power parties do not depend on the lower power parties (Magee & Galinsky, Reference Magee and Galinsky2008). For example, subordinates rely on supervisors for promotion opportunities; servers depend on customers to earn tips. On the other hand, higher power parties rarely depend on lower power parties to gain monetary resources. In addition, higher power parties tend to display a greater interpersonal variability in behaviors than the lower power parties, implying that their behaviors are less constrained by the environment. Hollander’s theory of idiosyncratic credits (Reference Hollander1958) suggests that people with higher power tend to be less confined by the social norm, thereby behaving as they please. Therefore, it appears that the more a target depends on a perpetrator for resources, the higher likelihood that misbehaviors by the higher power parties will be tolerated or accepted.
Furthermore, a target’s silence may be mistakenly taken as a sign of appropriateness for the misbehavior, which encourages the perpetrator to continue his or her behavior. Bargh and Raymond (Reference Bargh and Raymond1995) suggest that some perpetrators are unaware of the inappropriateness of their actions.
Practical recommendations
Following our argument that power and dependency are two critical aspects in understanding workplace sexual harassment, we make practical recommendations for designing better sexual harassment training in this section.
Clearly define the power boundary of higher power parties
In sexual harassment training, it is not only important to educate employees with clear definitions of inappropriate behaviors but also to inform them about the legitimate boundaries of authority held by those with greater power. This may begin by reinforming employees of their job descriptions and training them on the issues of power abuse with detailed examples. The purpose of doing so is to prevent potential victims, most likely the lower power employees, from tolerating or licensing misconduct by those with greater power. Although the people with high power have more legitimate authority to influence those with low power to a certain degree, it is essential for both parties to keep in mind the boundaries of their influence and understand that it is not their obligation to comply with commands beyond their job duties. Thus, including this aspect into sexual harassment training would potentially improve its effectiveness.
Loosen the dependency of lower power parties on higher power parties
Emerson (Reference Emerson1962) argues that a power dynamic represents the relative state of dependency between two parties. The extent to which A has power over B is largely determined by the extent to which B depends on A for certain resources and the availability of alternative suppliers of the same resource to which B can have access (Emerson, Reference Emerson1962). If B is in need of a valuable resource and can only obtain it from A, meaning this resource is important and there is a lack of alternative supply, we can say that B is highly dependent on A. In the organizational context, if an employee’s chance of career advancement is important to him/her, and is only determined by a direct supervisor’s evaluation, the employee may attempt to manage positive impressions and show a high compliance to the direct supervisor. The more dependent the employee’s career-advancing resources are on the supervisor, the more power the supervisor has over the employee. This further increases the likelihood of the supervisor exploiting the employee without facing much resistance.
Expanding availability of the resources given to those with lower power can potentially disconnect the power structure (Emerson, Reference Emerson1962). From the organizational and managerial perspective, it is essential to ensure that one’s career success is not fully contingent on decisions made by one, single, powerful person. We argue that it is important to assign multiple mentors to look after the lower-level employees and assign multiple raters to evaluate employees’ performances with standardized criteria. A multiple mentor system is not only beneficial to the fairness of performance appraisal but also essential to improving the effectiveness of sexual harassment training by intervening in the power and dependency structure in workplace.
Increase accesses for reporting incidents
Outside of the training context, it is also important to ensure that there is easy access to methods for the targets and bystanders to report sexual harassment incidents. The reports should be filed through easy formal procedures, such as filling out an online form to immediately report sexual harassment to the human resources department. If the report has to go through direct supervisors, the reporting system may not be as helpful, especially when a direct supervisor is the perpetrator.
Although it may be helpful to provide formal procedures for reporting sexual harassment incidents, individuals with lower power may be reluctant to utilize them if they fear retaliation or being ignored afterward. Therefore, perceived availability of informal procedures outside of the training context can be equally important in addressing the sexual harassment problem. Informal resources should include coaching, counseling, resolving harassment situations, and instrumental or emotional support by supervisors and coworkers (Berdahl & Raver, Reference Berdahl, Raver, Zedeck and Zedeck2011).
Fair performance appraisal and compensation
It is also critical that organizations ensure equal compensation to employees with the same job positions, regardless of their demographic backgrounds. In order to ensure such gender pay equality, organizations should accurately represent each employee’s merit in the compensation structure. Organizations should also ensure that raters of performance appraisal are not biased toward any demographic group, as both unfavorable and favorable biases may jeopardize the fair ratings (Harris, Reference Harris1994). When designing criteria for performance appraisal, organizations should ensure that the criteria are relevant to the job and are unbiased toward any group regardless of the level of power. Moreover, organizations are encouraged to provide a transparent compensation and promotion system to enhance the perception of fairness among their employees.
Conclusion
In conclusion, we argue that organizations could minimize the occurrence of sexual harassment and increase the effectiveness of sexual harassment training by understanding the power and dependence dynamic reflected by organizational structures. Moreover, ensuring that power relations are well-controlled in all environments outside training can facilitate enduring transfer effects from the training, further reducing sexual harassment in the long term.