Unfortunately, there is little evidence that training reduces sexual harassment. Rather, training programs, along with anti-harassment policies and reporting procedures, do more to shield employers from liability than to protect employees from harassment. And the clearest message they send is to the courts: Nothing to see here, folks.
There have been only a handful of empirical studies of sexual harassment training, and the research has not established that such training is effective. Some studies suggest that training may in fact backfire, reinforcing gendered stereotypes that place women at a disadvantage.
A 2001 study of a sexual harassment program for faculty and staff at a university found, based on responses to a questionnaire, that training increased knowledge about laws pertaining to sexual harassment but had no significant positive effects on behavior. Men who participated in the training were less likely to view coercion of a subordinate as sexual harassment, less willing to report harassment and more inclined to blame the victim than were women or men who had not gone through the training.
Forcing people into training may be especially ill-advised. Another study found that mandatory diversity training — which is broader than, but similar to, harassment training — did not increase the proportion of white women in management and actually led to a decrease in the proportion of black women. Voluntary diversity training had more positive effects.
A recent task force commissioned by the Equal Employment Opportunity Commission cautioned against drawing general conclusions from the small body of research, with each study looking at a differently designed training intervention. But real-world case studies suggest skepticism. For example, California has mandated sexual harassment training since 2005 for all supervisors at companies with more than 50 employees. And yet that didn't prevent widespread and egregious harassment at San Francisco-based Uber, the University of California system and the state Capitol, among other workplaces.
Likewise, virtually all large employers in the United States have anti-harassment policies and complaint procedures in place. Yet workplace sexual harassment remains common.
My research — based on surveys of organizations, interviews with human resources personnel and analyses of federal court opinions — has found that such policies coexist with work environments where prominent supervisors pressure subordinates for sexual encounters and where sexual banter makes women and others uncomfortable. (Workplace harassment on the basis of race, disabilities or religion is also common.)
Sexual harassment complaint procedures are often ineffective, either because they are not publicized or because the culture of the organization makes employees who experience harassment reluctant to take action. Multiple scholars have shown that employees often choose not to complain in order to show that they are team players, to avoid further humiliation, because they fear retaliation or because they think their complaints will not be taken seriously. Employees often prefer to think of themselves as survivors who can take harassment, as opposed to victims who cannot. Some worry that if they speak out, their powerful emotions will cause them to behave inappropriately, so they bottle up the humiliation. Some blame themselves. Many women (and people of color) have come to take harassment for granted as a part of workplace life that must be endured to succeed.
Interviews that my collaborators and I conducted with human resources professionals, as well as a study by Anna-Maria Marshall of sexual harassment in a university setting, have shown that those who handle complaints — usually HR professionals — often discourage potential complainants, suggesting that they should just ignore the harassment or that the behavior does not meet the (ambiguous) legal standard for sexual harassment.
When HR professionals do investigate, my interviews and analyses of human resources journals have shown, they often characterize sexual harassment complaints as instances of poor management or as interpersonal difficulties rather than as violations of law. To remedy the situation, they may transfer the complaining employee to another department or advise one or both parties to seek counseling or arrange for an apology, but they rarely take strong disciplinary actions against harassers. My interviews showed that employers may issue a warning and sometimes even impose a financial penalty, but public action (such as firing the harasser or revealing the reason for a penalty) is rare, because they worry about defamation suits by perpetrators as much as or more than discrimination suits by victims.
Title VII of the Civil Rights Act of 1964 bans discrimination on the basis of sex but does not explicitly address sexual harassment. Following the 1979 publication of Catharine MacKinnon's book "Sexual Harassment of Working Women," the Equal Employment Opportunity Commission, the federal agency responsible for enforcing the law, issued guidance advising employers to take preventive action against harassment. My surveys, as well as those of Frank Dobbin and Alexandra Kalev, show that employers, spurred largely by the human resources profession, quickly created written policies banning sexual harassment and complaint procedures for employees who had experienced it.
Human resources publications from that time show that there wasn't much concern for the potential targets of harassment. Instead, management scholars and HR officers claimed that these policies would protect employers from liability in the event that a supervisor harassed an employee.
They proved correct. In 1986, in Meritor Savings Bank v. Vinson , the Supreme Court recognized sexual harassment as a form of sex-based discrimination and suggested that an effective anti-harassment policy and a complaint procedure — one that is "calculated to encourage victims of harassment to come forward" — might protect an employer from liability. Twelve years later, with the support of an amicus brief by the Society of Human Resource Management, the Faragher v. City of Boca Raton and Burlington Industries, Inc. v. Ellerth cases explicitly created a defense that allows employers to escape liability for some types of sexual harassment by showing that they had taken steps to prevent and promptly respond to claims and that the employee had not taken advantage of remedial options provided by the employer. The decisions in those cases state that a written anti-harassment policy and an "effective" complaint procedure would normally meet the employer's legal burden to demonstrate preventive action.
Yet courts often fail to distinguish between meaningful compliance efforts and the merely symbolic policies and procedures that do little to protect employees' legal rights. Examining a representative sample of 1,188 federal court opinions from 1965 through 2014, my research collaborators and I found that courts tend to dismiss suits against employers who show that they have a policy banning harassment and a complaint procedure in place. Judges frequently discount evidence employees present showing that, despite having such policies, employers condone or fail to correct a culture in which harassment and degradation of women are common. And even where employees have good reasons for avoiding complaint procedures, such as fear of retaliation, courts generally see an employee's failure to complain as a bar to winning a sexual harassment case.
Consider, for example, Leopold v. Baccarat, Inc. , decided by a federal trial court in New York in 2000 and affirmed by a federal appeals court in 2001. Andree Leopold, a saleswoman for the Baccarat china and crystal company, sued after her supervisor repeatedly threatened to replace her with someone "young and sexy " and referred to saleswomen using vulgar, dismissive language. Baccarat pointed to its policy against harassment and its grievance procedure. Leopold claimed that she did not use the grievance procedure because (like many procedures) it would have required her to report the harassment to her immediate supervisor, who in this case was the supervisor she was accusing of harassment, and because the policy lacked a guarantee that she would be protected from retaliation. The court ruled in favor of Baccarat anyway, recognizing that the procedure was inadequate in some ways but nonetheless stating that "the law is very clear that any reasonable policy will do."
Studies in social psychology conducted by Cheryl Kaiser and Brenda Major may help explain judges' tendencies to disregard evidence of harassment when companies have anti-harassment policies and complaint procedures in place. In experimental work, Kaiser and Major have found that the presence of diversity efforts creates an "illusion of fairness," which causes people to overlook unequal treatment.
We have become a symbolic civil rights society, in which symbols of diversity and equal opportunity often mask legal violations.
So what might actually make a difference?
For our judicial system to provide effective redress, plaintiffs' lawyers must challenge ineffective anti-harassment policies and complaint procedures that are not accessible or fair. Judges, for their part, should avoid inferring a good-faith effort to prevent harassment from the mere presence of training, policies and complaint procedures, and they should recognize that employees often fear or are discouraged from using those complaint procedures.
Certainly there are some organizations in which managers take civil rights seriously and human resources personnel take great pains to investigate and remedy complaints of harassment. But employers who care about protecting potential targets of harassment, rather than just their own liability, should consider making anti-harassment training voluntary, not mandatory. They should make complaint systems accessible and ensure that discipline is prompt and proportionate. They should conduct climate surveys to assess the prevalence of harassment. And they should hold managers accountable for ensuring that their units are harassment-free — with compensation reflecting their success in doing so. The "me too" legislation introduced by Rep. Jackie Speier (D-Calif.) and Sen. Kirsten Gillibrand (D-N.Y.) on Wednesday includes some of these additional elements (although it still proposes mandatory training). The goal is to reform how Congress handles harassment complaints in a way that is more supportive of victims and less protective of the institution.
And yet we need not just rules and procedures but a broad recognition that power and inequality make it easy for people at the top to abuse people lower in the hierarchy, and extremely difficult for those at the bottom to do anything about it. A culture free of harassment will require widespread respect for women and equal representation of women in leadership. Capitol Hill is a good place to start.