SEXUAL HARASSMENT IN THE FEDERAL WORKPLACE Trends, Progress, Continuing Challenges A Report to the President and the Congress of the United States by the U.S. Merit Systems Protection Board

Ben L. Erdreich, Chairman

Beth S. Slavet, Vice Chairman

Antonio C. Amador, Member

Office of Policy and Evaluation

Director Evangeline W. Swift

Deputy Director John M. Palguta

Project Supervisor Bruce Mayor

Project Manager Karen K. Gard

Project Analyst Jamie J. Carlyle, Ph.D.

Sexual Harassment in the Federal Workplace
Trends, Progress and Continuing Challenges

Executive Summary

Background

Findings

Recommendations

1. Introduction and Background

Why This Study?

Sources of Information

A Note About Terminology in This Report

2. Defining Sexual Harassment: Changing Perspectives of Federal Workers

Broadening Definitions

Differing Definitions

Issues in Defining Sexual Harassment

Are More Precise Definitions Needed?

Are Ambiguous Definitions Better?

Intent Versus Impact

Concerns About Overemphasizing or Trivializing Sexual Harassment

3. Amount and Characteristics of Sexual Harassment in the Federal Sector

How Much Harassment Is Occurring?

Incidence Rates In Agencies

The Nature of Harassment

Types of Behaviors Experienced

Attempted or Actual Assault or Rape

Who Are the Targets of Sexual Harassment?

Who Are the Harassers?

Comparison With Non-Federal Organizations

Issues in Considering the Prevalence of Sexual Harassment

Behaviors Redefined

The Impact of Less Serious Behaviors

Mitigating Factors

4. Impact of Sexual Harassment

Monetary Costs to the Government

Computing Sexual Harassment Costs

Total Cost

Effects on Employees

5. Handling Sexual Harassment

Victim Reactions and Employee Voices

Response of Victims

Effectiveness of Victim Responses

Formal Action

Federal Supervisors’ Knowledge, Beliefs, and Response to Harassment

Are Supervisors Prepared to Handle Sexual Harassment?

Where Are Improvements Needed?

A Note on the Continuing Accountability of Supervisors

Agencies Address the Problem

Sexual Harassment Policies and Guidance

Sexual Harassment Training

Handling Complaints

Additional Measures Agencies Need to Consider

6. Court Decisions and Evolving Views of Sexual Harassment

EEOC Guidance

Supreme Court’s First Sexual Harassment Decision

Supreme Court’s Second Sexual Harassment Decision

7. Summary, Conclusions, and Recommendations

Conclusions

Recommendations

Appendix 1: Survey Instrument*

Appendix 2: Percentage of male and female respondents who said they had experienced the indicated unwanted behaviors in the previous 2 years

Appendix 3: Percentage of male and female respondents who said they had experienced the indicated unwanted behaviors once and more than once.

Appendix 4: Percentage of male and female respondents with the indicated characteristics, by victim and nonvictim status

Appendix 5: Percentages of male and female respondents who said they had taken the indicated action in response to unwanted sexual attention

Appendix 6: Percentage of victims who said that the indicated formal action made things better, made things worse, or made no difference

Executive Summary

In the 15 years since a concerned Congress first asked the U.S. Merit Systems Protection Board (MSPB or the Board to study sexual harassment in the Federal workplace, attention to sexual harassment issues has intensified. Two studies conducted by the Board in the 1980's found that sexual harassment in Federal offices and installations was widely perceived to be a problem. The questions and concerns that were being raised both within and outside the executive branch prompted the Board to undertake a follow up study to determine the nature and extent of sexual harassment in the Government today, to examine the actions Federal agencies have taken to address the problem, and to look at the pertinent issues through the eyes of Federal employees.

(It is important to note that this report uses the term sexual harassment" to characterize uninvited and unwelcome sexual attention and/or behavior reported by Federal employees, and that not all the conduct referred to as sexual harassment in the report would necessarily meet the more narrow legal definition of that term as established by legal opinions of the courts and the Board in the course of the past decade

This report presents the results of a Governmentwide survey of Federal workers who provided information on their attitudes and beliefs about relationships in the workplace, as well as data on their reported experiences with sexual harassment, the effects it had on them, and the programs agencies use to combat it. Where applicable, we have compared our findings with those from MSPB's 1980 and 1987 studies of this issue. We also looked at judicial developments and at the initiatives agencies described to prevent or eliminate uninvited, unwanted sexual attention in their organizations. The results indicate that while the Federal workforce, like society in general, is more sensitive to the issue of sexual harassment, the problem has by no means disappeared. Nevertheless, the Government has made progress in building a greater awareness of sexual harassment, a better understanding of the relevant issues, and increased sensitivity to the way people expect to be treated at work.

Background

While the Federal Government has done a great deal to address sexual harassment in the workplace, refining and refocusing of programs and policies are necessary to continue this progress and eliminate the problems that persist. Managers and supervisors need to make it clear, in actions as well as words, that they care about how their employees treat one another. Agencies need to identify their worst problems and best programs and tailor their future efforts accordingly. At the same time, agencies must be careful not to overreact to allegations of harassment or make assumptions about guilt or innocence before investigating the situation.

Because it costs taxpayers so much in terms of time lost, work disrupted, and legal battles engaged, sexual harassment makes victims of us all. As the workforce is reduced and agency budgets decrease, there is no corner of the Government wherein the Nation can afford to tolerate conduct that diminishes productivity, erodes morale, and directly conflicts with the standards of ethical behavior demanded of all employees.

Findings

In 1994, 44 percent of women and 19 percent of men responding to our survey reported that they had experienced some form of unwanted sexual attention during the preceding 2 years—rates similar to 1987's 42 percent and 14 percent.

The fact that the incidence of unwanted sexual attention has not decreased since the last Governmentwide survey is naturally a cause for concern. Despite very widespread training and information efforts that have successfully raised workforce sensitivity to the issues surrounding sexual harassment, the persistence of this amount of unwanted sexual attention in the Federal workplace suggests that the Government's programs to eradicate the problem need some serious reexamination.

At the same time, it is possible that at least some of this unwanted sexual attention was reported by survey respondents not in spite of efforts to increase awareness, but because of them. Individuals who formerly might have dismissed an uninvited look or remark, or persistent unwanted social invitations as mere rudeness or insensitivity, may now be more inclined to place those behaviors in one of the categories the Board's survey identifies as uninvited and unwanted sexual attention. (In fact, suggestive looks, sexual remarks, and employees pressuring co­workers for dates were the most frequently reported forms of sexual harassment, despite there being a number of respondents who said they would not characterize this conduct as sexual harassment.)

Formal responses, such as filing grievances or discrimination complaints are rare.

Only about 6 percent of the 1994 survey respondents who had experienced sexually harassing behaviors indicated that they took formal action in response to the harassment. Of the self-identified victims who did not take formal action, the most common reason (given by half these victims) was that they did not think the situation was serious enough to warrant such action.

Federal agencies have been successful in educating the workforce and raising awareness about sexual harassment.

Over 87 percent of Federal supervisors and 77 percent of nonsupervisory employees have received training in the area of sexual harassment. Some 78 percent of employees said that they know the channels to follow if they have been harassed and want to report it. All Federal agencies have policies prohibiting sexual harassment, and 92 percent of Federal employees are aware of those policies.

Sexual harassment cost the Federal Government an estimated $327 million during the 2-year period April 1992 to April 1994, but the overall ill effects of sexual harassment have decreased significantly.

This amount includes the cost of sick leave, job turnover, and productivity losses, and represents an increase since the Board's last sexual harassment study, when Government costs were estimated at $267 million for the period May 1985 to May 1987. However, the increase reflects inflation and the rise in salaries to a greater degree than it reflects an increase in the ill effects of harassment. Since the 1987 study, there has been a significant drop in turnover and sick leave used in response to sexual harassment, as well as a decline in the severity and duration of productivity losses resulting from the disruptive effects of sexual harassment.

The definition of sexual harassment is expanding, as more Federal employees are defining more kinds of behavior as sexual harassment.

Survey respondents were asked whether they would classify as sexual harassment six kinds of behavior, ranging from sexual comments to pressure for sexual favors. In virtually every case, whether the behavior was engaged in by a supervisor or by a coworker, the proportion of respondents—both men and women—who classified the six behaviors as sexual harassment rose between 1980 and 1987 and had increased again by 1994. Some of the increases are striking. For example, since the Board's first sexual harassment survey, the proportion of men who categorize uninvited sexual teasing, jokes, remarks, or questions by co­workers as sexual harassment rose from 42 percent in 1980 to 64 percent in 1994.

As in previous surveys, 1994 survey results show that the less severe forms of sexually harassing behaviors are the most prevalent, while the most severe behaviors occur the least often.

In 1994, 37 percent of women and 14 percent of men said they had experienced unwanted sexual teasing, jokes, remarks, or questions, generally considered less severe forms of sexual harassment. Actual or attempted rape or assault was reported by 4 percent of female respondents and 2 percent of males.

Coworkers and other employees, rather than individuals in the supervisory chain, continue to be the primary source of sexual harassment in the Federal workplace.

In 1994, some 79 percent of male and 77 percent of female respondents who reported experiencing sexual harassment said that they had been harassed by coworkers or other employees. This contrasts with the 14 percent of men and 28 percent of women reporting sexual harassment who said that an immediate or higher level supervisor had been responsible for the harassment.

Some employees are at greater risk than others of being targets of unwanted sexual attention.

Employees who have experienced unwanted sexual attention are more likely than those who have not experienced such attention to work exclusively or mostly with people of the opposite sex and to be supervised by members of the opposite sex. Employees of both sexes who reported having experienced unwanted sexual attention are more likely to be college-educated than those who have not experienced such attention. Also, employees under the age of 35 have a greater chance of experiencing unwanted sexual attention than those who are older. At the same time, the majority of employees who reported these experiences are 35 and older, since the population of employees in that age group is so large (83 percent of respondents).

The most effective responses to sexual harassment are informal, assertive ones such as confronting harassers and telling them to stop.

Although the most common response to unwanted sexual attention is to ignore the behavior or do nothing (44 percent of respondents who had experienced harassment reacted that way), asking or telling a person to stop was identified by 88 percent of all survey respondents as the action they believed would be most effective in dealing with harassment. Of the respondents who had actually experienced sexual harassment and taken this action, 60 percent said it made things better.

A sizable number of employees, particularly men, are concerned about how they will be perceived by others in the workplace, in view of today's emphasis on sexual harassment.

Some 63 percent of the male and half of the female respondents indicated that they believe that some people are too quick to take offense when someone expresses a personal interest in them through looks or remarks. One in three employees believes that normal attraction between people in the workplace is, to a moderate or great extent, misinterpreted as sexual harassment. Nearly half the men indicated they don't feel comfortable giving compliments because their remarks might be misinterpreted.

Most employees do not think that the emphasis on sexual harassment has made their workplaces uncomfortable.

Employees' concern about how others perceive their words and actions may be causing people to think more critically about the effects of their conduct and to exercise more self restraint, but it apparently has not led to a chilling effect in the workplace. Only 18 percent of men and 6 percent of women indicated that fear of being accused of sexual harassment had made their organizations uncomfortable places in which to work.

Comments provided by survey respondents indicate that some perceive the penalties for harassment to be inappropriate or inconsistent.

While it may be the case that most supervisors and managers want to stop harassment in their organizations, some may prefer to do it in a way that avoids harming the career of the harasser, who otherwise may be very valuable to the organization. Some survey respondents provided comments indicating that they see this as resulting in penalties that are too light or that demonstrate a double standard, with higher-level or managerial and executive personnel being treated less harshly than lower level employees.

Most Federal agencies have not diagnosed the nature and extent of sexual harassment within their own organizations and sub elements.

Because agencies, for the most part, have not identified their worst trouble spots, programs to educate the workforce and to eliminate sexual harassment tend to be generic, aimed at the agency's entire workforce, rather than targeted towards specific problem areas.

Recommendations

1. Agencies should find ways to capitalize on what is already known about the most effective actions that can be taken to prevent and eliminate sexual harassment; that is, they should publicize penalties and encourage assertive actions on the part of employees who are targets of unwanted sexual attention.

The results of all three MSPB sexual harassment surveys have shown that employees believe that publicizing sexual harassment policies and penalties are among the most effective actions agencies can take to prevent sexual harassment. The nearly universal awareness of sexual harassment policies among members of the Federal workforce indicates that agencies have done a good job in getting the word out about their policies. Less is known among the workforce about what happens to people who harass others. Employees should be made aware of how the agency intends to discipline proven harassers. Victims should always be informed about what happened to their harassers, and penalties should be public enough to serve as examples to potential harassers that management's prohibition of sexual harassment is more than lip service.

As indicated earlier, the most effective approach for targets of unwanted sexual attention is to take assertive actions such as confronting harassers and telling them to stop, or reporting the behavior to someone in a position to help. Agencies should facilitate this approach by highlighting assertiveness in their training programs and by making it easier for victims to report harassing behaviors through informal programs such as neutral advisors or an ombudsman who serves as a confidential consultant to victims.

2. Managers and supervisors should be firm and consistent in penalizing harassers.

When harassment occurs, managers and supervisors should take action based on the serious­ness of the offense rather than the rank of the offender. In deciding a reasonable penalty to be imposed when harassment has been proven, managers and supervisors should not give undue weight to the harasser's performance and value to the agency. Managers and supervisors must understand that the value of a harasser's contributions to the organization is likely to be diminished by behavior that hurts morale, demonstrates a lack of ethics, or exhibits a double standard. Further, the example that management sets in following through with appropriate penalties can be more effective as a preventive measure than the policies it promulgates.

3. Agencies should diagnose the extent and seriousness of sexual harassment within their own organizations so that they know what kinds of solutions are appropriate and where resources should be concentrated.

The content and goals of agency programs to eliminate sexual harassment should be linked directly to what is known about the nature of sexual harassment in the agency. Studies and surveys that help agency policymakers see the work environment through employees' eyes can help in devising remedies that are sensitive to an agency's multiple cultures; e.g., headquarters activities, field activities, administrative operations, health care facilities, law enforcement operations, scientific laboratories. Knowing what and where the most serious problems are can help agencies target scarce energy and resources in the most efficient ways. As much as sexual harassment costs the taxpayer, and as lean as future agency budgets are likely to be, Federal organizations cannot afford to direct insufficient attention to serious problems while expending resources in areas where problems are minimal or nonexistent.

4. Agencies should evaluate the effectiveness of the sexual harassment training they provide to ensure it addresses identified problems. Agencies should pay particular attention in their training efforts to the problem of sexual harassment by coworkers.

Sexual harassment training is provided in every agency at all organizational levels. This training represents a considerable investment, and while most agencies know whether or not it's popular with participants, they generally don't know what kind works best, what parts of it are effective, what kinds make no difference, and whether any of it has a negative effect. Therefore, agencies should adapt their training to address what they learn from their own self-diagnoses of the extent of sexual harassment in their workforces and from studies such as this one. For example, because findings consistently show that coworkers and other employees are the primary source of sexual harassment in Federal agencies, training efforts might emphasize strategies for handling harassment from peers. Follow ups should be conducted to determine what effect, if any, training actually has on the targeted workforce, and training content should be revised if it is found to make no appreciable difference in preventing or stopping sexual harassment. Agencies must also ensure that important programs such as sexual harassment training be given the proper type and degree of emphasis, in keeping with what is known about the nature and extent of the problem in their own organizations.

CHAPTER 1

Introduction and Background

Why This Study?

It has been over 15 years since a concerned Congress first asked the U.S. Merit Systems Protection Board to examine the nature, amount, and impact of sexual harassment in the Federal workplace. Both that initial examination in 1980 and a follow up study conducted in 1987 indicated the widespread incidence of sexual harassment in Federal offices and installations, and revealed rates of harassment that remained virtually unchanged in the years between the first and second studies.

In the years since 1987, attention to sexual harassment issues has continued almost unabated and, in fact, intensified greatly in the early 1990's, when sexual harassment incidents in Federal agencies and charges against Federal managers and nominees to high office became headline news. From the private sector—by no means immune from these problems—came equally unsettling reports of sexual harassment in venues as wide-ranging as factories, law offices, universities, medical establishments, and high schools.

The public's level of awareness, if not anxiety, had definitely been raised. And both within and outside the Federal executive branch, a number of questions and concerns surfaced. For example, in an enterprise such as the Federal Government, with its much-heralded emphasis on equality, fairness, and employee protections, how could such unsavory (not to mention illegal) events continue to occur? Hadn't the Government done enough to prevent such abuses? What do we know about the extent and nature of sexual harassment in the Federal workplace today? Are we channeling human and fiscal resources properly in attempting to prevent and treat the problem?

Given the questions that were being raised, and the Board's own ongoing commitment to reviewing and advising on this issue, it was time to look at the problem again. As the independent agency responsible under the law for studying how the U.S. civil service system assures the health of the merit systems and the absence of prohibited personnel practices in Government, the Merit Systems Protection Board took on, in early 1994, its third examination of sexual harassment in the Federal workplace. Our purpose was to determine the incidence of sexual harassment in today's Federal workplace, to review the actions taken by Federal agencies to eliminate it, and to look at the pertinent issues through the eyes of Federal employees.

Sources of Information

Over 61 percent of those who received the survey returned completed questionnaires, giving us more than 8,000 returns. Nearly 1,700 of those respondents provided additional written comments regarding their answers to survey items and their opinions and beliefs about sexual harassment. The responses came from employees from all the cabinet-level departments and a number of the largest independent Federal agencies--22 departments and agencies in all—as well as from workers employed by a cross section of smaller Federal organizations which we refer to in this report as "other agencies. "2

The survey results represent the experiences and opinions of nearly 1.7 million permanent civilian employees in the executive branch of the Federal Government. The workforce represented by survey respondents is about 57 percent male and 43 percent female, and includes both blue-collar and white-collar employees at grade levels 1 through 15, and members of the Senior Executive Service.

Both supervisors and nonsupervisors were surveyed, and respondents represent all ages (18 and above) and educational levels.

In addition to the survey, our study included a review of relevant literature and background discussions with several focus groups. One group comprised Federal Women's Program managers and EEO officials from several agencies, and two groups were made up of a cross-section of Federal employees who provided feedback on the survey instrument and shared their thoughts about the causes and effects of sexual harassment.

We also sent a set of questions to Federal agencies requesting information on their efforts to combat sexual harassment. (These are the 22 departments and agencies listed in footnote 2.)

A Note About Terminology in This Report

It's important to note, in considering the meaning of the term "sexual harassment" in this report, that not all the behaviors that we are calling harassment, and that Federal workers identify as sexual harassment in our survey, would necessarily qualify as sex discrimination in a legal sense. The behaviors described may include instances of offensive conduct, not necessarily pervasive or extreme, that Federal workers find unacceptable but that are not necessarily cause for legal action.

But focusing exclusively on sexual harassment so extreme as to meet a legal test was never the aim of the Government's information and prevention programs. In confronting the issue of sexual harassment, the Federal Government is interested not only in avoiding situations in which a court would find a violation of law, but also in preventing the creation of an unpleasant, unproductive work atmosphere. The sexually harassing behaviors reported by survey respondents and discussed in this report—whether or not they are cause for legal action—can most definitely create an unproductive working environment and thus are an appropriate focus of our attention.

It should also be noted that when the term "harassment" is used in this report, it refers to sexual harassment, not any other type of harassment.

CHAPTER 2

Defining Sexual Harassment:

Changing Perspectives of Federal Workers

The way Federal employees define sexual harassment is one of the issues the Board has continued to monitor since the first administration of our survey in 1980. We have found that the views of Federal employees about what is and is not sexual harassment, while not completely uniform, are becoming more alike in the sense that more people of both sexes have come to view more behaviors as sexual harassment.

Broadening Definitions

For the third time since 1980 we described, in our survey, six kinds of behavior, ranging from sexual remarks to pressure for sexual favors, and asked respondents whether they would consider the behavior sexual harassment if engaged in by a supervisor and if engaged in by a coworker. These behaviors are:

• Uninvited letters, telephone calls, or materials of a sexual nature
• Uninvited and deliberate touching, leaning over, cornering, or pinching
• Uninvited sexually suggestive looks or gestures
• Uninvited pressure for sexual favors
• Uninvited pressure for dates
• Uninvited sexual teasing, jokes, remarks or questions

In virtually every case, the proportion of respondents—both men and women—who classified the behaviors as sexual harassment rose between 1980 and 1987, and had increased again by 1994. For some behaviors, the change from 1980 has been dramatic. For example, the percentage of men who believe that a coworker pressuring someone in the workgroup for sexual favors is sexual harassment rose from 65 percent in 1980 to 93 percent in 1994. Likewise, the percentage of men who said they consider a coworker's uninvited sexual remarks to someone in the work-group to be sexual harassment rose from 42 to 64 percent.

A similar pattern is present for women. As might be expected, in response to all three of our sexual harassment surveys, a consistently higher proportion of women than men classified all behaviors on our list as sexual harassment. However, there was still room for the numbers of women who view the behaviors as sexual harassment to rise, and the 1994 survey responses show that they did just that. For example, the percentage of women who consider coworkers' sexual remarks to be sexual harassment increased from 54 percent of respondents in 1980, to 64 percent in 1987, to 77 percent in 1994. We found similar increases for some of the other less serious behaviors on the list. For the more serious behaviors, such as pressure for sexual favors, the proportion of women classifying the conduct as sexual harassment was already high.

There is no doubt that people today are interpreting what happens in the workplace differently from the way they did in the 1980's. The offensive comment or off color story that might have been tolerated in the 1980 workplace may in the 1995 environment be reinterpreted as suggestive speech, and be categorized and reported as an incidence of sexual harassment. Boorish and discomfiting behavior that in the past might have been accepted as the price of keeping a job is no longer considered by most employees an unavoidable part of earning a living. The 1994 survey results (see table 1) seem to bear this out.

Table 1
Is It Sexual Harassment? *

There's general agreement among Federal employees of both sexes that all the behaviors we listed are sexual harassment. None of the behaviors was classified as sexual harassment by less than 64 percent of respondents to the 1994 survey. In the case of pressure for sexual favors, virtually all men and women consider the behavior sexual harassment. Likewise for deliberate touching, although fewer men than women think that when a coworker (as opposed to a supervisor) does it, it's sexual harassment (89 percent for men, versus 96 percent for women).

This fact illustrates a point about the workforce's attitude towards unwelcome behaviors engaged in by coworkers compared with supervisors. In this latest administration of the sexual harassment survey, just as in the previous ones, respondents appear to be holding supervisors to a higher standard than coworkers. For every one of the behaviors we listed, respondents were more likely to define a behavior as sexual harassment if a supervisor does it than if a coworker does it.

Differing Definitions

Despite the increasing likelihood that Federal employees will agree that the behaviors listed in our survey constitute sexual harassment, there are still some behaviors that a number of people do not agree on. For example, while growing numbers of people consider a coworker's sexual remarks to be sexual harassment, more than one in every five men and one in every eight women responding to our 1994 survey said that such remarks are not sexual harassment.

In addition, there was a marked degree of uncertainty among some respondents about how to classify some conduct, especially when coworkers are the source of the unwelcome behavior. One in ten employees responded that they don't know whether suggestive looks and pressure for dates by coworkers are sexual harassment. More than one in eight employees responded that they don't know whether coworkers' sexual jokes or remarks are sexual harassment. In all cases where uncertainty was an issue, more men than women indicated they are unsure about the behaviors.

Not surprisingly, when the responses indicated disagreement among survey participants or indecision about how to classify a behavior, the behavior in question invariably was one of those generally considered less serious. Thus, suggestive looks, sexual remarks, and employees pressuring coworkers for dates remain, in the minds of many Federal employees, the most ambiguous among the behaviors addressed in the survey. They are also, as discussed later in this report, the most frequently occurring unwelcome behaviors in the Federal workplace.

Issues in Defining Sexual Harassment

In our examination of how Federal employees define sexual harassment, several issues emerged as the ones evoking the most concern. These include the perceived need for a more precise definition of sexual harassment, the desirability of letting the definition remain ambiguous, the notion that people's intentions are what count in deciding whether they've harassed someone, and concern about the effects of making the definition of sexual harassment too broad. These issues are discussed below.

Are More Precise Definitions Needed?

A number of comments from focus group participants and from survey respondents assert that the Government needs to do a better job at defining exactly what sexual harassment is. "Someone needs to develop a definitive explanation of what does and does not constitute sexual harassment," one respondent wrote. Another said, "The subtle forms [of harassment] are difficult to judge and prove—many violators are clueless that these are harassment. It's those gray areas that need to be more clearly defined for both the victim and the prospective perpetrator, so little doubt will exist that the victim has an undeniable reason to issue a complaint."

While the desire for better definitions is understandable, it may not be achievable. As the Supreme Court's 1993 decision in Harris v. Forklift Systems, Inc., suggests about the precision of sexual harassment definitions, "This is not, and by its nature cannot be, a mathematically precise test."3 (See ch. 6 of this report for a discussion of the Harris case.) Thus, the people who wish for a fixed, detailed definition of sexual harassment may continue to be disappointed and those gray areas may remain so. With sexual harassment, as with other types of discriminatory behavior, it is nearly impossible to enumerate all of the potentially inappropriate actions that could possibly fit a general definition.

Are Ambiguous Definitions Better?

Although some survey respondents want sexual harassment more strictly and consistently defined, others see a positive side to allowing harassment to be defined by the situation. According to one respondent, "[T]here seem to be so many shades of gray in interpersonal relationships that I really have trouble with black and white definitions. Maybe if we'd give each other a little more compassion, respect, and understanding these great definition hunts wouldn't be necessary.”

Other respondents indicated a preference for ambiguous definitions for somewhat different reasons. They fear that strict definitions—with forbidden topics and actions clearly enumerated—could stifle the relationships and behaviors that foster productivity in the workplace. As one respondent wrote, "This is a touchy subject because of the different views each person holds about what is acceptable. I don't think it is necessary or desirable to prohibit all actions that the most conservative person would find offensive. There would be no interaction with coworkers on a personal level, and this would cause a decline in communication, teamwork, and productivity."

A number of respondents noted that they worry about people misclassifying some behaviors as harassment. In written comments to us as well as in survey items, Federal employees expressed concern about how workers perceive—and might misperceive—one another's conduct. For example, we asked employees how they feel about complimenting the appearance of others at work. Nearly half the men (though only 14 percent of the women) who responded indicated that they don't feel comfortable giving compliments because they might be misinterpreted. In another question relating to interpretations of behavior we asked respondents about people's reactions when someone expresses a personal interest in them. Men and women were in much closer agreement on this item, with 63 percent of the men and half of the women indicating that they believe some people are too quick to take offense when someone expresses a personal interest in them through looks or remarks. One in every three respondents indicated their belief that normal attraction between people is, to a moderate or great extent, misinterpreted as sexual harassment.

What these results suggest is that people are thinking critically about how their conduct is perceived by others, and perhaps modifying it accordingly. This is not a bad outcome of the Government's focus on sexual harassment. If employees realize that a compliment or an expression of personal interest might offend some people, they might make a habit of critically judging how their words or deeds will be interpreted by others before they speak or act. A heightened sensitivity to how one's fellow employees perceive the world is not a recipe for disharmony in the workplace, despite the complaint of some workers that it seems riskier to give a compliment than it used to be. It is not, after all, a hardship to simply refrain from giving a compliment if one is not sure how it will be received.

But has this apparent anxiety among Federal employees about how people are going to interpret one another actually led to disunity and unhappiness in the workplace? Not that we can tell. In fact, the response to one of our survey items seems to refute a frequently heard claim that all this attention to sexual harassment has had a chilling effect in the workplace. Only 18 percent of men and 6 percent of women respondents agreed that fear of being accused of sexual harassment had made their organizations uncomfortable places to work. Apparently the increasingly acknowledged need for self-restraint doesn't necessarily equate to discomfort on the job.

Intent Versus Impact.

Some of our survey participants insisted that an action or behavior should be considered sexual harassment only if the individual engaging in it intends harm. As one respondent put it, "To define sexual harassment in terms of someone's perception rather than in terms of objective actions is absurd. What is gained by categorizing sociocultural or interpersonal misunderstandings as criminal behavior?"

There's an understandable reluctance to accuse someone of sexual harassment who meant no harm. Some employees are simply oblivious to the impact their behavior has on others. And people do make mistakes. But perhaps because in this context the behavior has a sexual overtone, there is less tolerance for such "mistakes" than there would be for errors made about less sensitive matters. In fact, the majority of our survey respondents didn't seem to agree that benign intentions should be the deciding factor in whether conduct is judged as harassment. Over 54 percent of men and 59 percent of women agreed that there are certain behaviors they would consider sexual harassment even if the person doing them did not mean to be offensive.

Concerns About Overemphasizing or Trivializing Sexual Harassment.

Many participants in our survey expressed a concern that labeling too many things as sexual harassment could obscure the larger problems. One commenter said, "If there are too many hypersensitive people crying foul over the most innocuous comments, no one is going to take the serious complaints seriously." These fears may or may not be well-founded. It's probably not particularly helpful to those employees who do suffer serious harassment for every instance of ill-mannered behavior to be construed as sexual harassment; people can become inured to the truly egregious cases of sexual harassment that do occur.

Nevertheless, while respondents expressed concern about elevating minor offenses beyond their importance, our survey does not provide evidence that the increased attention given to sexual harassment in recent years actually has resulted in minimizing or ignoring charges of serious harassment. We do know that the majority of the women (64 percent) and a significant proportion of the men (43 percent) whom we surveyed do not think that too much attention has been paid to the issue of sexual harassment in the past several years. (See table 2.)

Table 2
1994 Survey Item: "Too much attention has been paid to the issue of sexual harassment in the past several years.

At the same time, however, nearly one in six women and almost one in three men in the Federal workforce do believe that too much attention has been paid to the issue. Moreover, many of the survey respondents who provided written comments expressed concern about overemphasis on sexual harassment or about the level of resources the Government devotes to sexual harassment issues.

Managers should not ignore the concerns of employees who believe that the sexual harassment issue has been blown out of proportion. The emphasis given to important programs such as sexual harassment prevention has to be kept in proper balance. If Federal agencies don't assure that their managers and supervisors give these programs the right type and degree of emphasis, the unintended results can be harmful.

A number of survey respondents who wrote comments about this issue provided anecdotes that illustrate the kinds of negative effects that can result when program emphasis is out of balance.

Some people indicated that some managers in their agencies appear to focus on sexual harassment primarily out of a desire to be politically correct. These officials are seen as lacking in genuine concern about the problem and its victims. Targets of sexual harassment who sense this apparent hypocrisy may be reluctant to report incidents of harassment or may fear retaliation from their harasser. Potential harassers who get the idea that management isn't really serious about dealing with sexual harassment severely, may feel they can misbehave with impunity.

In either case, the work and resources the Government devotes to eliminating sexual harassment can be undermined if employees feel the effort is insincere. Agency leaders should be conscious of this problem and should assure that attention paid to sexual harassment in the form of policies and training programs are not (and are not perceived as) mere lip service or a substitute for taking actions that produce real improvements in the workplace.

CHAPTER 3

Amount and Characteristics of Sexual Harassment in the Federal Sector

How Much Harassment Is Occurring?

For well over a decade, sexual harassment has been a highly visible issue in the Federal Government. Official policies forbidding sexual harassment, training about what sexual harassment is and how to handle it, and campaigns to raise employees' awareness of its perniciousness have contributed to a broadening sensitivity about the problem and its effects.

Given the amount of attention that has been and continues to be paid to sexual harassment, it comes as rather a surprise that in 14 years and three administrations of our sexual harassment survey, we have seen no decline in the rate of sexual harassment reported by survey participants. Some 44 percent of the women and 19 percent of the men who responded to our survey in 1994 reported having experienced harassing behaviors during the preceding 2 years. Comparable figures for 1980 and 1987 are 42 percent for women in both years and 15 percent (1980) and 14 percent (1987) for men.

As in the previous surveys, to determine the extent of sexual harassment in the workplace we asked workers if, during the preceding 24 months, they had received unwanted attention in any of these forms:

1. Actual or attempted rape or assault
2. Pressure for sexual favors
3. Deliberate touching, leaning over, cornering, or pinching
4. Sexual looks or gestures
5. Letters, telephone calls, or materials of a sexual nature
6. Pressure for dates
7. Sexual teasing, jokes, remarks, or questions

In the 1994 survey we included an additional form of harassment—stalking—which we defined as "unwanted following or intrusion into your personal life."4 The overall incidence rates for all these forms of sexually harassing behavior are depicted in figure 1.

Figure 1
How Many Employees Experienced Sexually Harassing Behaviors in the Previous Two Years?

Incidence Rates in Agencies.

The extent of sexual harassment in each individual agency is normally a matter of special interest to department and agency managers. Our survey data indicate that for nearly all agencies the proportion of employees reporting sexual harassment over the 2-year period preceding the 1994 survey rose compared to earlier reporting periods. (See table 3.)

Table 3
How Much Harassment Is Occurring?
Percentage of employees in 1980, 1987, and 1994 who reported experiencing sexual harassment, by agency

Note on agencies in this table. In 1980, the Board did not collect data from the State Department, the Small Business Administration, NASA, or the Office of Personnel Management. The Department of Health, Education and Welfare was listed in the 1980 survey, but after the survey was developed the agency was abolished and the Departments of Education and of Health and Human Services were formed. The 1994 figures for the Department of State do not include the U.S. Information Agency or the Agency for International Development; the 1987 figures did include these agencies in the Department of State. The category of other DOD" includes agencies other than the military services, e.g., the Defense Logistics Agency, the Defense Mapping Agency, and the Office of the Secretary of Defense. All agency figures are based on responses from employees at worksites in the continental United States, Alaska, Hawaii, and Puerto Rico.

The Nature of Harassment

Types of Behaviors Experienced.

By far the most commonly experienced harassing behavior reported by our survey respondents is unwanted sexual teasing, jokes, remarks, or questions. Nearly 37 percent of women and 14 percent of men reported experiencing this sort of verbally harassing behavior. For both male and female employees, this is also the only one of the unwanted behaviors that has shown a slight but steady increase at each administration of the survey.

The other very commonly occurring behaviors that survey respondents experienced are un­wanted sexual looks or gestures and unwanted touching, leaning over, cornering, or pinching. The least common harassing behaviors reported by respondents are actual or attempted assault or rape and pressure for sexual favors. (See table 4.)

Table 4
Forms of Sexual Harassment
Percentage of respondents who experienced the indicated behaviors during the preceding 2 years

Overall, the changes from previous surveys in the percentages of respondents reporting each form of harassment have been quite small, usually no more than two percentage points, if there is any change at all. (Details and comparisons with previous surveys can be found in app. 2.)

Our 1994 data also show that the less serious forms of harassment are not typically one-time occurrences. The majority of both male and female victims of these behaviors reported experiencing them more than once. For the more serious kinds of harassing behaviors, the data indicate that about as many victims had the experience once as had it multiple times. (See app. 3.)

For a sizable number of survey respondents, the unwanted attention they experienced went on for a month or more. Some 55 percent of respondents reported that their experiences with harassing behaviors lasted from 1 month to more than 6 months. For about 35 percent of those who reported being harassed, the unwanted attention went on for less than a week. The duration of the harassment reported by 1994 survey respondents is about the same as that reported by participants in the 1987 survey.

Attempted or Actual Assault or Rape.

Perhaps some of the most unsettling statistics coming from the 1994 survey are those involving attempted or actual rape or assault. The percentage of women who reported having been subjected to attempted or actual rape or assault rose from 0.8 percent in 1987 to 4 percent in 1994. Of male respondents, 2 percent reported being victims of this behavior, an increase from 0.3 percent in 1987.

It should be borne in mind that these data include assault as well as rape, and attempted as well as actual occurrences of both. The experiences reported can include a number of behaviors, from shoving to actual, forcible rape. In other words, a range of behaviors from serious to extremely serious may be included in this category of behavior.

Who Are the Targets of Sexual Harassment?

By analyzing demographic information provided by respondents to our 1994 survey, we found that the characteristics of respondents who said they had experienced unwelcome sexual behaviors differ in several respects from those who did not report those experiences (see app. 4):

• Victims are more likely than nonvictims to work exclusively or mostly with individuals of the opposite sex.
  
• Victims of both sexes are more likely than nonvictims to be unmarried.
  
• Victims are more likely to be supervised by members of the opposite sex than nonvictims.
  
• Both male and female victims are more likely than nonvictims to have attended college.
  
• Employees under 35 have a greater chance of being harassed than those who are older. The proportion of employees in the under 35 age group who have been harassed is larger than the proportion of victims in the 35-and-older age group. For example, 56 percent of female respondents who are under 35 reported experiencing unwanted sexual attention, in contrast to 42 percent of female respondents who are age 35 and older. At the same time, because the population of employees who are 35 and older is so large (83 percent of respondents are in that age group), the majority of victims are 35 and older.

Typical Victims of Harassing Behaviors

In addition to looking at the demographic characteristics of sexual harassment victims, we wanted to find out if their attitudes towards relationships in the workplace or to harassment-related issues differed from those of respondents who had not experienced unwelcome sexual attention.

For the most part, victims' beliefs about personal relationships and behavior in the workplace mirrored those of nonvictims. On a few issues, however, there was as much as a 10-percentage-point difference between the two groups.

Among male respondents, we found a difference in the way victims and nonvictims view sexual joking or conversations about sexual issues. While over 61 percent of the males who had been targets of sexual harassment said this kind of talk is almost always inappropriate in the workplace, half of the nonvictims expressed this belief.

A similar degree of difference appeared between male nonvictims' and victims' attitudes towards witnessing the harassment of others. Among men who had not experienced unwanted sexual attention, about 24 percent agreed with the statement "I would consider myself a victim of sexual harassment if I witnessed someone else in the workplace subjected to unwanted sexual attention." In contrast, 34 percent of the men who had experienced sexually harassing behavior expressed agreement that witnessing sexual harassment would make them victims of sexual harassment.

On one item, differences appeared between victims and nonvictims for both male and female respondents. We asked whether respondents would consider certain behaviors to be sexual harassment even if the person doing them did not mean to be offensive. Among both men and women, just over half (around 52 percent) of nonvictims agreed that they would consider some behaviors to be sexual harassment notwithstanding the intentions of the offender. Among victims of sexual harassment, a significantly larger proportion--65 percent of the men and 68 percent of the women—agreed that intention was irrelevant.

These results suggest that sensitivity to sexual comments or behavior in the workplace is higher among respondents who reported experiencing sexually harassing behaviors in the two years preceding the survey. What we do not know is whether the victims' attitudes had formed before the experiences they reported on the survey or their attitudes developed as a result of the experiences.

Who Are the Harassers?

Although sexual harassment of a relatively powerless subordinate by a more powerful supervisor is what most people picture when they think of sexual harassment, responses to this survey as well as the Board's previous sexual harassment surveys indicate that harassment by coworkers is far more common. And, as would be expected, unwanted sexual attention usually comes from members of the opposite sex.

Sex of the harasser. Responses to our 1994 survey indicate that among those who have experienced unwanted sexual attention, most males (65 percent) have been harassed by women and the overwhelming majority of females (93 percent) have been harassed by men. About 1 percent of women victims said they'd been sexually bothered by other women, while a significant number of male victims (21 percent) said that other men had harassed them. (The other sources of harassment were mixed groups of men and women or unknown sources, as in the case of anonymous letters.)

These data indicate that the sources of harassment by sex of the offender have not changed very much since the 1980 survey. Respondents to the 1994 survey reported somewhat more harassment that is anonymous or done by both males and females. We do not know how much same sex harassment is perpetrated by heterosexuals and how much by homosexuals. We do know from written comments provided by the respondents that both kinds occur.

Organizational relationship between harasser and victim. By far the most likely sources of unwanted sexual attention were persons other than the supervisors of the victims. About 79 percent of male victims and 77 percent of female victims were subjected to unwanted behaviors by people they identified as coworkers or other employees without supervisory authority over them. Some 14 percent of male and 28 percent of female victims were sexually harassed by persons in their supervisory chains.

As shown in table 5, the organizational source of sexual harassment has changed relatively little since the Board's last sexual harassment survey.

Table 5
Who Are the Harassers?
Percentage of victims sexually harassed by supervisors and others

Perspectives of employees accused of sexual harassment. We asked respondents whether they had ever been accused of sexually harassing someone, and what their feelings were about the situation. Very few employees responded to these questions, but of those who did, most had been accused of sexual harassment by coworkers or subordinates. The vast majority of these respondents did not believe that the complaints against them were fair. Most believed that they had done nothing wrong, the complainants had misunderstood their motives, and/or the complainants wanted to make trouble.

Comparison With Non-Federal Organizations

Sexual harassment is just as critical a topic outside the Federal sector as inside. No field of endeavor is immune. But whether there is more harassment inside or outside the Federal Government is an issue on which survey respondents who have held jobs in both places have varying opinions.

Twenty-two percent of survey respondents who have worked outside the Federal Government believe there is more harassment outside the Government and 7 percent said there's less. The proportion of respondents who said there's the same amount of harassment within and outside the Government is 34 percent, but the largest single group--36 percent—indicated they didn't know or couldn't judge. Results of the 1987 survey were similar, with 20 percent of respondents indicating there was more, and 8 percent indicating there was less, sexual harassment outside the Federal sector. A larger proportion of respondents at that time (42 percent) thought the amount of harassment was the same.

Issues in Considering the Prevalence of Sexual Harassment

Confining the examination of sexual harassment to a review and periodic comparison of incidence rates does not provide a complete picture of a topic as complex as this one. The various sources we referred to in examining this subject, including the results of our survey and the comments of survey participants, raise several issues that should be considered in order to permit a fuller understanding of the problem. These are discussed below.

Behaviors Redefined.

As is evident from table 3, there have been few agencies where the proportion of employees who experienced sexual harassment has declined. In the majority of agencies, the 1994 percentages increased from 1987, sometimes doubling or more. However, the increases shown for 1994 may, to some extent, reflect the increased awareness of sexual harassment and the broadening definitions of harassment that we have seen Federal workers adopt. Because so many workers are now sensitized to the issue, it's possible that people who formerly would have dismissed an uninvited look or remark as mere rudeness may now be more inclined to place that behavior in one of the categories the survey identifies as uninvited and unwanted sexual attention.

Information provided for this study by Federal departments and agencies lends support to this hypothesis. In response to a question we asked agencies about whether sexual harassment is more or less of a problem than it was 5 years ago, half the agencies said that requests for counseling and reports of sexually harassing behavior had increased. At the same time, however, they observed that the increases appeared to be related to the growing public awareness of the problem. A number of agencies also contended that their training and prevention efforts along with the considerable amount of attention focused on sexual harassment has made employees more willing to come forward with complaints.

The Impact of Less Serious Behaviors.

The question has arisen since the Board's last report on sexual harassment as to whether the percentage of employees experiencing sexual harassment primarily reflects a high proportion of individuals who only experience the "less serious" forms of sexually harassing behaviors. To test this theory, we calculated incidence rates excluding these less serious forms of harassment (i.e., looks/gestures, pressure for dates, letters/calls, and jokes/remarks) when respondents reported they'd experienced any of these only once. Our assumption was that if any of these acts occurred just once they were less likely to be considered harassment than if the behavior were repeated.) Nonetheless, we found the percentage of employees who reported experiencing some form of sexual harassment other than those less serious behaviors was still fairly high--38 percent for women and 15 percent for men. These data, then, suggest that the percentages of employees who reported experiencing sexual harassment are not due primarily to the inclusion of isolated incidents of bad manners or poor judgment in our calculations of the extent of sexual harassment.

At the same time, it's worth noting that over 90 percent of the men and women who reported experiencing harassing behaviors said that they did not take formal action in response. And of that group, half reported that they took no formal action because the offense wasn't serious enough. Therefore, while many acts of offensive conduct are uninvited and unwanted, and appropriately may be characterized as sexually harassing behavior, quite a few of the employees who are tar­gets of that behavior appear to find it, if not inconsequential, at least not a matter for a special response.

Mitigating Factors.

Somewhat mitigating the fact that the percentage of employees who reported experiencing sexual harassment in the Federal workplace has held steady are our survey results on how employees react to harassment. As discussed in the next chapter, the sexually harassing behaviors that respondents reported being subjected to did not usually cause them to use annual or sick leave or leave without pay.

In fact, comparing the 1994 results with those of the 1987 survey reveals that a smaller percentage of victims reported taking any type of leave as a result of sexually harassing behaviors. For example, only 8 percent of victims responding to the 1994 survey had used sick leave, contrasted with 13 percent in 1987.

The next chapter further explores the costs associated with employees' reactions to behaviors they consider sexually harassing.

CHAPTER 4

Impact of Sexual Harassment

Whether seen through the eyes of the victim or from the coworker's or the agency's perspective, there is no doubt that sexual harassment has had a serious and sustained impact on the Federal Government. For employees who experience it, sexual harassment takes its toll in the form of mental and emotional `stress and even loss of income, if victims leave their jobs or take leave without pay as a result of their experiences. For the Government as an employer, the dollar costs attributable to lost productivity and sick leave are very high.

Monetary Costs to the Government

In 1980 the Board estimated that for the 2 years preceding that year's survey, sexual harassment cost the Federal Government $189 million. For the 2 years preceding the 1987 survey, the cost of sexual harassment was estimated at $267 million. Our most recent figures, covering the 2 years preceding administration of the 1994 survey (April 1992 to April 1994), show an estimated cost to the Government of $327 million.

The 1994 estimate represents an increase over the cost figures derived from 1987 study. However, this increase reflects inflation and the rise in salaries to a greater degree than it reflects an increase in the ill effects of harassment. Although a larger number of 1994 respondents who had experienced sexually harassing behaviors reported a decline in productivity than did our 1987 respondents, the amount of the decline was less. Further, as indicated earlier, fewer respondents reported leaving their jobs or using sick leave because of harassment than in 1987. Nevertheless, the price that employees and the Government pay in reacting to and dealing with sexual harassment is far too high.

Computing Sexual Harassment Costs.

In computing the cost of sexual harassment in the Federal Government we take into account the cost of job turnover, sick leave that victims say they used as a result of the harassment, the cost of the individual productivity decreases reported by victims, and the estimated productivity lost by work groups in which harassment occurs. These elements are generally computed separately for men and women (because their average annual salaries differ) and then are added together for a dollar total.

Our estimate of the cost of sexual harassment is conservative. Among the items we did not include are the cost of benefits paid by the Government and the cost of overtime for other workers who fill in for employees absent because of the effects of workplace harassment. Nor did we factor in the cost of dealing with informal complaints, processing formal ones, and handling litigation. A summary of the factors used to arrive at the estimate follows.

Job turnover. Based on data provided by our survey respondents regarding how sexual harassment affected them, we estimate that in the 2 years preceding the 1994 survey, sexual harassment caused 19,727 Federal employees (victims) to leave their jobs through reassignment, being fired, being transferred, or quitting. This is a decrease since the Board's last sexual harassment survey, which found that an estimated 36,647 employees had left their jobs because of their experiences with sexually harassing behaviors. Although the population represented by the survey respondents has decreased by 16 percent since 1987, the turnover decrease since 1987 is over 46 percent.

The expenses associated with replacing these employees include the cost of offering jobs to the replacements (recruitment and placement costs); the cost of background checks for new or potential employees; and the cost of training the replacements. Turnover estimates for the 1987 study conservatively set replacement costs at $1,000 per employee. Increasing that amount— again conservatively—to account for inflation, we estimate that employees who left because of harassment in the 2 years preceding the 1994 survey cost $1,250 each to replace. The price of turnover among Federal employees, then, amounted to an estimated $24.7 million during the 2-year period of the study.

This amount is 33 percent lower than the $36.7 million turnover cost for the period preceding the Board's 1987 survey. The turnover among men is down 60 percent and the turnover among women is 39 percent lower than in 1987.

Sick leave. The emotional and physical impact that sexual harassment has on its victims comes with a high price tag for the employer as well as for the employee. About 8 percent of survey respondents who had experienced harassment reported using sick leave as a result. As with job turnover, the use of sick leave as a response to sexual harassment has shown a significant decrease since the Board's 1987 study, when 13 percent of individuals who had experienced unwanted sexual attention reported using sick leave as a result.

In arriving at the cost of harassment-related sick leave, we used responses of survey participants to calculate the total number of hours of sick leave used by men and by women Governmentwide as a result of sexual harassment. Using those estimates and the average annual adjusted basic pay for male and female Federal employees, we calculated a total sick leave cost of about $14.9 million.5 This represents an 87 percent drop in sick leave usage among men and a 35 percent drop for women, with a consequent 43 percent reduction since 1987 in the cost of sick leave usage resulting from sexual harassment.

Individual productivity losses. Even when the targets of sexual harassment do not find it necessary to use leave as a result of their experiences, some report that the amount or quality of their work suffers during and following the experiences. Survey respondents who had experienced harassing behaviors were asked to indicate how much, if at all, their productivity had been reduced as a result of the unwanted attention and how long the reduction continued.

Although nearly 9 out of 10 victims indicated that they suffered no reduction in productivity, or only a slight loss, the total effect on work quantity and quality and the dollar value associated with the reduction are still considerable. In determining how much the loss of productivity cost the Government, we took into account these responses along with the average annual adjusted pay for males and females, and arrived at an estimated total of $93.7 million for the 2-year period preceding the survey.

As with turnover and sick leave usage, these results reflect a lessening of the negative effects of sexual harassment since the Board's 1987 study. Although the dollar cost of individual productivity losses was higher in 1994 than in 1987 (when it was estimated to be $76.3 million), the higher figure reflects higher salaries in 1994 rather than more time lost due to the disruptive effects of sexual harassment. While in 1994 a larger percentage of respondents reported a loss of productivity, the loss wasn't as severe and didn't last as long as it did for 1987 survey respondents. We estimate that the amount of time lost due to sexual harassment has declined by about 37 percent for men and stayed about the same for women.

Work group productivity losses. Because sexual harassment can affect not only the individual victims but also their coworkers, supervisors, and others with whom they interact at work, work group productivity is included in our estimate of harassment's cost to the Government.

For the study period covered by the 1987 survey, the cost of work group productivity losses was estimated at over $128 million. These costs were calculated on the basis of a survey question which asked employees who had experienced unwanted sexual attention whether the unwanted attention had affected the productivity of others in their work group. Factoring in the rise in average basic pay for men and women since the last survey, we estimate the cost of work group productivity losses for the 1994 study period to be $193.8 million.

Total Cost.

Our estimate of the cost of sexual harassment to the Government over the 2-year period for which victims were reporting may be summarized as follows:

If these cost figures seem too large to have much down-to-earth meaning, it may help to equate lost time to lost money. For example, imagine an employee who's being bothered by a coworker who leers at her or makes comments full of innuendo or double entendres, or who tells jokes that are simply inappropriate in a work setting. The time this employee spends worrying about the coworker, the time she spends confiding in her office mate about the latest off-color remark, the time she spends walking the long way to the photocopier to avoid passing his desk, is all time that sexual harassment steals from all of us who pay taxes.

Adding up those minutes and multiplying by weeks and months begins to paint a picture of how costly sexual harassment is. Increase this one individual's lost time by the thousands of cases like this in a year, and the waste begins to look enormous. And this may well be a case that doesn't even come close to being considered illegal discrimination by the courts. Whether or not they're illegal, these situations are expensive.

Effects on Employees

In addition to the substantial dollar amounts sexual harassment costs the Federal Government, there are very real and sometime severe costs— both financial and emotional—borne by the employees who experience unwanted sexual attention.

In the 2 years covered by the 1994 survey, Federal employees who took leave without pay because of sexual harassment lost wages estimated at $4.4 million. The estimated amount of annual leave that victims used during the period totals over 973,000 hours. Some victims reported that they quit or were transferred or even fired because of sexual harassment. (See table 6.)

Table 6
What Is Sexual Harassment's Impact on Victims?
Percentage of respondents who experienced sexual harassment and took or experienced the indicated action, 1987 and 1994

At the same time, as noted above, there's been a decline since 1987 in the amount of turnover, the severity and duration of productivity losses, and the proportion of people calling in sick or taking other kinds of leave as a result of harassment (although some victims' reluctance to miss work may reflect a general national nervousness about holding on to one's job in an environment in which layoffs are becoming more common and fewer jobs are readily available outside the Government).

Nevertheless, when these negative situations result from sexual harassment, the consequences can be devastating to the individual victims. Here is how several respondents described their experiences:

As a result of my complaint, I [was] ostracized by the group and [was] the topic of idle gossip. It became an all-consuming issue. It carried over at home where my family also suffered in helping me deal with it.

I was very upset by his request for a sexual favor. My superior performance rating was lowered by him to fully acceptable. I did not want to hurt his career, but it hurt mine. I felt I must resign. After six months on unemployment, which was very degrading, I returned to work with the government, having to take a downgrade. This experience has left me very bitter and down on myself and my abilities

We saw this person harass several female employees over the years he was there. He was even the reason that one employee was fired when he harassed her so badly that she was no longer able to keep the sustained average required in her position.

Sexual harassment makes victims of the recipients of unwanted attention, their coworkers, and the agencies where it occurs. And, although there has not been a big jump in the rate at which the cost of sexual harassment has risen since our last study of the issue, these taxpayer dollars do not represent an investment with a healthy return. This is money lost and in some cases, damage permanently done.

CHAPTER 5

Handling Sexual Harassment

The problem of how to eliminate sexual harassment in the Federal workplace has inspired a great many actions, programs, and potential solutions. While there may be isolated exceptions, as a rule, everyone in Government has heard about the issue, and most have an opinion regarding what should or shouldn't be done about it.

The efficacy and attractiveness of the potential solutions to the sexual harassment problem depend on one's perspective as an employee, a victim, a supervisor, or a management official. Our survey data and information we gathered from Federal agencies about their programs provide insights into these varying perspectives.

Victim Reactions and Employee Voices

Response of Victims.

The range of responses for a victim of sexually harassing behavior is probably as vast as the range of human behavior itself. The actions, interactions, and relationships that give rise to behavior that is or can become harassing are remarkably complicated. Human beings say things that belie their actions and do things that belie their words. And for some, it's very difficult to know what to do, how to say whatever is necessary to stop unwanted sexual attention, and how to predict what will happen as a result.

Perhaps that is why the most frequently occurring reaction to sexual harassment is inaction. The single most common response of employees who are targets of sexually harassing behaviors hasn't changed since the initial administration of MSPB's survey in 1980. That response has been, and continues to be, to ignore the behavior or do nothing. In 1994, about 44 percent of victims indicated that they reacted this way, with men and women equally likely to do so. The reason for some of this inaction may be related to the insignificance of the offense; many people who are targets of harassing behavior do not find it worth bothering about. But there are some victims whose experiences with unwanted sexual attention are quite serious, and they still do nothing.

The other most common reactions to unwanted sexual attention are asking or telling the harasser to stop and avoiding the harasser. Table 7 lists the most common actions that victims who responded to our survey reported taking. Appendix 5 shows details on the responses of males and females to harassing behaviors.

Table 7
How Did Victims React?
Percentage of victims who said they took the indicated informal action in response to sexual harassment, 1994

Effectiveness of Victim Responses.

As was found in previous MSPB sexual harassment surveys, some of the actions more likely to be taken by victims are not necessarily the most successful in putting a stop to the harassment. For example, of the 44 percent of victims who said in the 1994 survey that they had ignored the behavior or done nothing about it, only 22 percent reported that this had "made things better."

The majority of victims who ignored the unwanted behavior, went along with it, or made a joke of it, found that their actions (or lack thereof) made no difference in their situations. In fact, going along with the behavior seems to be about the least effective thing a victim can do. This is particularly true in the case of female victims, more than a third of whom reported that this response actually "made things worse."

If the more diffident responses to unwanted attention don't do much to curtail sexual harassment, what kinds of responses will help people who are faced with it? From the perspective of our survey respondents as a whole (both those who had experienced sexual harassment in the preceding 2 years and those who had not), three actions stood out as likely to be the most effective in stopping harassing behaviors:

• Asking or telling the person(s) to stop;
• Reporting the behavior to the supervisor or other official(s); and
• Filing a formal complaint.

Table 8 lists these and other actions and shows the percentage of survey respondents who said they believe that the action would be effective.

Table 8
What Should Targets of Sexual Harassment Do?
Percentage of all 1994 respondents who believe the indicated action would be most effective in stopping sexual harassment

But what actually worked for the survey participants who have experienced sexual harassment? In 1980, 1987, and again in 1994 the answer to that question was the same. And it's fairly consistent with the views of each year's respondents as a whole. The more assertive actions—such as confronting harassers by telling or asking them to stop—invariably were the responses that more victims reported had improved the situation. Victims found that reporting the problem to a super­visor or telling someone else also was more helpful than not. The only less assertive action that nevertheless seemed to make things better was, not surprisingly, avoiding the person doing the harassing. This, of course, can make things better by removing the victim from the source of the problem (although it probably won't help the harasser's next victim very much, should there be one). However, it can also have a negative effect on the victim's work performance, if she or he spends a lot of time trying to avoid the harasser. Table 9 provides information about the effectiveness of various actions actually taken by victims.

Table 9
How Well Did the Informal Actions Work?
Percentage of victims who said in 1994 that the indicated action made things better, made things worse, or made no difference

Many of the comments provided by respondents to our survey support the finding that two actions—confronting the harasser or reporting the situation to a supervisor or other official—are the best approaches for employees who are faced with this problem. Individuals who were able to stop the offensive behavior by confronting their harassers or by reporting the behavior to someone in a position to help did this in widely varying ways, from the barely assertive to the unequivocally forceful. Here is what some of them told us:

After speaking to the person and explaining how his actions bothered me, he understood, apologized, and has not done it since.

The person who harassed me does not work for the Federal Government; he works for the onsite contractor. I reported the incidents to management who were quick to address the problem and would have taken further action had it continued.

I had not gone to my supervisor in the beginning, [but] that ended up being the correct route in this situation. He put a stop to it as soon as I made him aware of the situation.

I basically told her that her advances were not welcome and that stopped her dead in her tracks.

I made it unpleasant for the person to talk to me or behave inappropriately by loudly saying back so that others could hear "What did you say?" I also told the person how I felt, and the behavior stopped after 2 or 3 weeks. I guess the point is, if you don't stick up for yourself, you are pretty much thrown to the dogs.

I slapped my supervisor in a room full of people after he [whispered] a very sexually explicit remark. Luckily, this happened at my going away party.

In reviewing respondents' comments, we found that while some people have no problem defending themselves or reporting harassers, for others it is difficult; for still others, nearly impossible. The respondents themselves, both victims and nonvictims, recognized and identified this as a problem:

The harassment was unwanted touching and remarks done blatantly in front of others. When one of the other trainees mentioned at an office party that I was being harassed I denied it was occurring and said I didn't mind, even though I did. I was afraid of telling anyone how I really felt. More emphasis should be placed on each person's responsibility to tell the harasser that his or her actions are not wanted

We have a good sexual harassment policy, but it fails to mention the role of the person being harassed to formally and immediately notify [the harasser]. I think if a person is immediately notified of inappropriate behavior, it could help solve several problems and improve communication between the sexes.

The success that targets of sexual harassment have had with assertive responses to their harassers, and the limited use to which this approach has been put deserve our attention. It may well be that people who find it difficult to confront a harasser will need help in adopting an assertive approach to dealing with offensive behavior.

Some might object—quite logically—that it is not the victims who should have to change their interpersonal style; the offenders should be the ones to change. But as a practical matter, if employees who experience unwanted sexual attention want to stop harassment in the workplace and get on with their jobs, the most expedient way is often to assertively put a stop to the misbehavior, by speaking out or reporting the offense, rather than waiting for harassers to see the error of their ways. Further, there are people in the workplace who are guilty of offensive behavior but totally unaware of it, and they need to be told. The experience of this survey respondent illustrates the point:

When I first started working for the Government, my immediate supervisor asked me out for a date every day for six months (I said no every day.) I was a nervous wreck because I couldn't understand why he persisted. Finally I told him very bluntly that I was not interested in pursuing a romantic relationship with him, and he stopped. The point is that I think he had no idea how much stress this caused me or that he was doing anything inappropriate.

Employees need to understand their own responsibility for dealing with harassment—by unequivocally rejecting the inappropriate behavior or, if this is impossible or proves to be ineffective, by enlisting the help of a supervisor or other official in dealing with the problem. And, of course, managers and supervisors need to foster a work environment that makes assertive solutions a natural choice for victims.

Formal Action.

One of the more assertive responses available to employees who experience harassment is that of taking some type of formal action. But not many victims actually do this. While 78 percent of all survey respondents (and 76 percent of victims) said they know the formal complaint channels available for sexual harassment victims, only about 6 percent of victims who responded to our 1994 survey said they had taken formal action.

These findings are consistent with the relatively small number of formal sexual harassment complaints filed with the U.S. Equal Employment Opportunity Commission (EEOC). In fiscal year 1993, 1608 allegations of sexual harassment were filed by Federal workers, a number that had significantly increased from the previous year, when 947 such complaints were filed. Nevertheless, the FY 1993 figure represents only about 3 percent of formal allegations of discrimination filed with the EEOC by Federal Government employees that year.6

Our 1994 survey results indicate that the kind of action most often taken by employees who did take formal action is requesting an investigation by the employing organization. Table 10 lists the kinds of formal actions employees took in response to unwanted sexual attention.

Table 10
What Kinds of Formal Actions Did Victims Take?
Percentage of 1994 respondents who experienced sexual harassment and took the indicated formal actions in response

We asked victims who said they had taken formal actions to rate the effectiveness of the action(s) they had chosen. Requesting an investigation by an outside organization, while apparently the least popular formal action, turned out to have been the most effective: 61 percent of victims who took this action said it made things better. Of those who requested an investigation by their employing agency, 47 percent found that the action made things better.

Filing grievances or discrimination complaints seems to have been a poor choice for many victims. Grievances or adverse action appeals helped 32 percent of the victims who filed them, but made things worse for 47 percent of victims who filed them7 Only 21 percent of victims who filed discrimination complaints found that it made things better, while 37 percent reported that it made things worse. (See appendix 6.)

We also asked survey respondents who did not take formal action to choose from a list of reasons why they did not do so (see table 11). The most common reason, given by half of the victims, was that they didn't think the offense was serious enough to warrant formal action. This is not surprising considering that the most common forms of sexual harassment (suggestive remarks or looks) tend to be the least serious. These results are similar to those from the Board's 1987 survey.

Table 11
Why Are Victims of Sexual Harassment Reluctant to Take Formal Action?
Percentage of victims who chose the indicated reason for not taking formal action in response to unwanted sexual attention, 1994

There is good news in the fairly low percentage of victims (6 percent) who said that they didn't take formal action because they didn't think their supervisors would be supportive, and in the percentage (5 percent) who professed not to know how to take formal action. But what should give Federal managers and supervisors cause for concern are those victims who indicated that they did not take formal action because they thought nothing would be done; they believed the situation would not be kept confidential; or they thought formal action would adversely affect their careers. Even if management can't control every member of its workforce and prevent the commission of every offensive act, doing the right thing after the harassment has occurred is within management's control. Therefore, it is extremely important for Federal managers and supervisors to be sensitive to the needs and working environment of employees who report harassment and be supportive of the actions they decide to take.

Aside from the specific reasons selected by sexual harassment victims who responded to the survey, there may be some elements inherent in the formal complaint process that make taking formal action a less desirable alternative than one of the other typical responses to unwanted sexual attention. For example, the lack of control victims have over the outcome of grievances or discrimination complaints may deter some from formalizing their actions. If a victim's primary focus is simply on getting the harasser to stop the offensive behavior, there may be little incentive to take an action that is more complicated, or an action that has objectives beyond stopping the behavior (punishment, for example), or one that may include outcomes not intended by the victim, such as damaging the harasser's career. (Not wanting to hurt the person who had bothered them is, after all, among the reasons given by 17 percent of victims for not wanting to take formal action.)

As one researcher has noted:8

Typically, the goal of informal processes is to end the harassment of the complainant rather than judge (and punish, if appropriate) the offender. The focus is on what will happen in the future between the disputing parties, rather than on what has happened in the past * * *.

The informal process typically ends when the complainant is satisfied (or decides to drop the complaint); the formal procedure ends when the hearing board decides on the guilt or innocence of the alleged harasser. Thus, control over the outcome usually rests with the complainant in the case of informal mechanisms, and with the official governance body in the case of a hearing * * *.

Thus, a desire to put uncomfortable situations behind them may be what moves many victims of sexual harassment to deal with it as they do. Informal methods of dealing with harassment, whether or not they ultimately result in punishing the harasser, appear to have the highest success rate and produce the highest comfort level among employees who experience unwanted sexual advances. As the approach of choice among victims, informal methods of dealing with harassment should be encouraged by supervisors, who should look for ways to facilitate their use by employees.

Federal Supervisors' Knowledge, Beliefs, and Response to Harassment

No matter how well victims themselves manage to handle unwanted sexual attention, if they are not consistently supported by their supervisors, the energy they expend in defending themselves, and the resources their agencies expend on addressing sexual harassment are in danger of being wasted. It's extremely important for supervisors and managers to set the right tone in their organizations, to let employees know what's expected of them in terms of how they treat one another, and to take decisive action to correct behavioral problems.

Supervisors' obligation under Title VII of the Civil Rights Act of 1964 to maintain a workplace free of discrimination in any form means that they have to make it their business to be sensitive to the atmosphere in their organizations and stop any behavior that could be considered harassing. It's essential that supervisors be trained in recognizing and dealing with sexual harassment. They must be aware of their agency and Government policies with regard to harassment, and they need to know the avenues open to employees to obtain relief when confronted with harassment.

Are Supervisors Prepared to Handle Sexual Harassment?

Based on their responses to our survey, Federal supervisors as a group appear to be well-trained in sexual harassment issues, and generally positive towards agency policies prohibiting harassment and the programs designed to deal with it.

Some 87 percent of supervisors who responded to our 1994 survey reported that they had received sexual harassment awareness training, most of them relatively recently. Over 88 percent of those trained had had sexual harassment training within the 2 years preceding the survey, and 70 percent reported being made more sensitive about sexual harassment issues or more aware of others' feelings as a result.

In the opinion of over 85 percent of supervisors, their employing agencies are making efforts to reduce or prevent sexual harassment "to a great or moderate extent." Most of them—almost 9 out of 10—also said that they know the formal complaint channels available to victims of sexual harassment in their agencies, and 71 percent cited publicizing the availability of formal channels as among the most effective actions an organization can take to curtail sexual harassment. Most also believe in the effectiveness of establishing and publicizing agency policies on sexual harassment (84 percent), and in universal (as opposed to supervisors-only) training on the subject (78 percent).

Where Are Improvements Needed?

Survey results suggest a corps of Federal supervisors who are informed about sexual harassment issues and aware of agency policies and complaint procedures. Further, very few survey respondents who had experienced sexual harassment cited lack of supervisory support as a reason for not taking formal action in response to their experiences. Nevertheless, data collected from survey respondents, as well as a number of their written comments, shed light on areas where some managers and supervisors need to improve the way they deal with sexual harassment when it occurs in their organizations. These supervisory and managerial shortcomings seem to fall into three main categories: being reluctant to confront harassers; taking inappropriate or inadequate actions against harassers; and failing to investigate or making errors in pursuing investigations of alleged harassment. These are discussed below.

Confronting harassers. Reluctance of supervisors and managers to confront harassers parallels the lack of assertiveness that prevents victims from confronting their harassers. In both cases embarrassment may be a factor: ours is not a society in which it is the norm for people to easily and comfortably discuss sex or sexual matters, especially in a setting in which sex normally is not a natural topic. Sexual joking or workplace conversations about sex are thought to be inappropriate by 54 percent of nonsupervisory respondents to our survey and a slightly larger number of supervisors (58 percent). (See table 12.) Feelings of unease or uncertainty about discussing sexual matters could certainly contribute to a lack of communication on those occasions when it would be appropriate to talk about sex-related subjects, as when discussing harassment. One of our respondents described a situation that illustrates the point:

I did not report it. The man involved, by all appearances, was a gentlemanly, grandfatherly sort. I think most people would not have believed me because they never saw that side of him. I also felt my supervisor would resent me for forcing him to counsel the employee—which I think would have embarrassed him. I think my supervisor was aware of some of it, but avoided it so he didn't have to talk to the guy.

This situation represents another very common workplace issue; that is, people's great difficulty in criticizing others, whether or not the criticism is warranted, and whether or not it is one's duty to offer the criticism (as in the case of supervisors criticizing their subordinates). The general dread that people feel about telling other people difficult truths that may provoke denial, anger, or embarrassment may also explain why supervisors don't readily confront harassers.

Table 12
How Do Employees Feel About Sexual Talk at Work?
1994 Survey Item: Sexual joking or conversations in which people talk about sexual issues are almost always inappropriate in the workplace.

Another factor that may contribute to the reluctance of managers and supervisors to confront harassers is an underlying belief that if anything sexual is involved the situation is none of their business:

One reason for the continuing problem is that supervisors often consider employees' social or sexual behavior toward each other as an issue outside their supervisory responsibility * * * the "It's not my job syndrome."9

Whether or not this is a common feeling among Federal supervisors, it remains their duty to make inappropriate interactions between employees in the workplace their business, particularly when those interactions are having an impact on the work of the organization. And not only must supervisors and managers be sensitive to trouble among their employees, they must step in to stop or prevent it. A survey respondent who had experienced harassing behaviors put it simply:

Management and supervisors have to be willing to counsel employees about their decorum. They should intervene instead of discount what an employee feels or perceives because they don't feel or perceive the same thing. I wish [my supervisor] had taken it upon himself to pay attention to the touching, etc., and correct this man.

Sanctions against harassers. Comments provided by a number of survey respondents suggest that while most supervisors and managers want to stop the harassment that they become aware of, many prefer to do it in a way that avoids harming the career of the harasser, who may, aside from the harassment, be a good employee and valuable to the organization.

Respondents to the Board's survey complained that higher level employees were the least likely to receive appropriate punishment for harassment, and that some managers and supervisors cover up or ignore complaints against accused harassers, particularly those who are themselves managers or supervisors. Here is what some survey participants said:

The higher the position of the man, the less is done to him—if anything at all. Complaints can be brought against the high-ranking person but at worst he gets a slap on the hand and that is the example that is set.

I was intimidated by my supervisor, stalked and threatened. When an IG investigation revealed/ proved his guilt, he was told he shouldn't have done it, don't do it anymore. End of story. I transferred to another agency.

I believe that no one trusts the harassment process to work when they only see the harasser moved around in the agency and sometimes even promoted! Each agency needs a management team that truly does punish the harasser and allows the victim to know how and to what extent the harasser was punished.

If a civilian reports harassment to a military's supervisor, the supervisor always says it will be handled—but it is never reflected in the military officer's performance report. The standard phrase in this office when there's an opening is, "Let's hire the next [employee] according to her bra size. " Generals put up with this because they don't want to hurt the officer's career.

Of the small proportion of victims who took formal action against their harassers (6 percent), 44 percent of the men and 22 percent of the women said that management's response was to do nothing. To the extent that this is also a common management response when formal action is not involved (when, for example, a victim informally reports harassing conduct to a supervisor), it is an unfortunate, short-sighted approach to handling problems among employees. Managers and supervisors must consider the fact that the value of the harasser's contributions to the organization is likely to be much diminished by behavior that erodes morale, demonstrates a lack of ethics, or exhibits a double standard.

Penalties for sexual harassment must be severe enough to convince employees that management is serious about stopping sexual harassment and public enough to deter offensive behavior in others. Managers and supervisors who have a role in judging and disciplining harassers must avoid leaving the impression that the conduct is condoned or minimized or swept under the carpet to avoid embarrassing the parties involved.

Investigating allegations. There are economic, legal, social, moral, and ethical incentives for supervisors and managers to take action to deal with allegations of sexual harassment. All these motives can be pretty compelling for the conscientious supervisor to whom a complaint is made about someone's behavior. But supervisors and managers have to be cautious not to overreact to allegations of harassment and make assumptions about guilt or innocence before investigating the situation. Fear of knee-jerk reactions on the part of management was another concern expressed by a number of our survey participants:

I am a woman manager. [M]ale colleagues of mine have suffered career damage because of false sexual harassment charges. Once the charge is made the defendant must prove his innocence. We must maintain perspective [and] a sense of fairness for all. I believe sexual harassment occurs—I experienced my share of it in the past. But I also know it is an easy charge for a vindictive person to make.

I have seen a case where no decision about guilt or innocence was made but the accuser was awarded with a better job. Management appeared to be taking the path of least resistance.

Law and policies require that harassers be dealt with, but do not compel an employer to fire them nor to assume that an employee accused of harassment is guilty until proven innocent. Investigations of alleged incidents need to be fair, thorough, and prompt, and it's extremely important that agencies ensure that their managers and supervisors know how to conduct fact-based investigations of sexual harassment charges. Punishments of persons found guilty of sexual harassment should fit the offenses and should be consistently applied. And it is equally important to punish accusers if investigations reveal that they knowingly made false allegations.

A Note on the Continuing Accountability of Supervisors.

While it is critical that supervisors be held accountable for taking actions to prevent sexual harassment and for putting a stop to it when it occurs, it is equally important for the Government to ensure that supervisors are prepared to deal with these sensitive issues. Managers shouldn't be surprised to encounter the kinds of people problems described by some of our survey respondents if they put individuals into supervisory jobs whose ability to manage a staff is questionable or whose interest in and sensitivity to the needs of a staff are minimal.

Some of these problems can be alleviated through supervisory training, but perhaps the best way to attack inadequate supervisory response to problems like sexual harassment is to focus on making better selections for supervisory jobs. Individuals who are expected to manage or prevent problems such as sexual harassment in the workplace must have the special skills required for such a task. As earlier Board reports have noted, however, Federal agencies often assign individuals to supervisory jobs based on their abilities in their technical fields, as opposed to their ability to manage people.10

As the Federal Government shrinks and organizations become flatter, selecting individuals for supervisory positions becomes even more critical. Therefore, if departments and agencies are to deal effectively with the problem of sexual harassment in the workforce, they must take into consideration the special abilities of their supervisory and management personnel. Ideally, in filling supervisory jobs, management will aim to assign individuals whose skill at managing people matches their technical excellence.

Agencies Address the Problem

Sexual Harassment Policies and Guidance.

In both the private and the public sectors, policies prohibiting sexual harassment are the norm. Human Resources Executive, a magazine that periodically polls executives in the human resources field, found in its 1993 executive census of over 300 human resource directors (from organizations ranging in size from under 500 to over 50,000 employees) that of all the policies and programs implemented by American companies, sexual harassment policies were by far the most common. Some 95 percent of firms had instituted sexual harassment policies at the time of the study. (In contrast, the next most popular initiatives were smoking cessation and employee assistance programs, which had been established, respectively, by 74 percent and 70 percent of the companies polled.)11

In the Federal workplace formal sexual harassment policies are even more common. Every one of the 22 major Federal departments and agencies we contacted in connection with this study has an established sexual harassment policy. All of these policies are current—none had been updated more than 2 years prior to our inquiry. Agencies typically reissue their policies whenever new agency heads assume office. In a few instances agencies update their sexual harassment policies annually.

Most of the agencies reported that the heads of their subordinate organizations issue their own policies locally. In four of the agencies that participated in this study, local policies are a requirement levied by the agency head. This practice can give policy statements more force because employees associate the policy with someone they actually know, recognize, respect, or can identify as a person with direct influence over their jobs.

The policies themselves vary in the amount of detail included and in the amount of emphasis given specifically to sexual harassment as distinguished from other issues related to equal employment opportunity. Some of the policies that cover sexual harassment are brief, general EEO policy statements that include sexual harassment as a form of illegal discrimination. Other, more detailed policies focus exclusively on sexual harassment and address definitions, responsibilities of all concerned parties, and avenues of redress for victims. A few agencies, such as the Department of the Navy, have issued more elaborate and detailed policies that feature instructions on recognizing and dealing with sexual harassment.

Policy effectiveness. In the opinion of our 1994 survey participants, agency policies do have an impact. Over 80 percent of the respondents counted establishing and publicizing sexual harassment policies among the most effective actions an organization can take to reduce or prevent sexual harassment. (Table 13 shows how employees rate agency policies and other measures, in terms of what the employees believe would be helpful.)

Table 13
What Do Employees View as Effective Preventive Measures for Agencies to Take Against Harassment?
Percentage of all 1994 survey respondents who believe that the indicated action would be among the most effective an organization could take

At the same time, somewhat fewer respondents--68 percent—said that their agencies' policies make a difference in the way people behave towards one another. (See table 14.) Although survey data indicate that most Federal employees view agency sexual harassment policies as effective, there is still a sizable minority who believe otherwise. It may be that formal policies need to be better implemented through practical programs in the workplace, and the actions of supervisors and managers need to reflect more faithfully the sentiments expressed in the policies.

Table 14
What Affects Sexual Harassment in the Workplace?
Percentage of 1994 respondents who gave the indicated responses

Whatever practical difference they make in peoples' lives, agency policies are almost as important for what they represent as for what they actually say or do. The fact that the realities of the workplace don't always reflect a policy's good intentions does not by any means render official agency policies useless. The policies are evidence that agency leaders are on record as intending to deal appropriately with sexual harassment. That stated commitment from the top can be critical in backing up managers and supervisors at all levels who are trying to foster a workplace environment in which sexual harassment is not tolerated.

Getting the word out. Most agencies responding to our questions described publicity programs, some of them quite elaborate, to inform all their employees about their policies to combat sexual harassment and to provide guidance to supervisors and employees on dealing with sexual harassment.

Methods of getting the word out to employees about agency sexual harassment policies range from sending a personal copy of the written policy to each employee through channels— about half of the agencies do this—to running radio and television stories about the policies (Defense Department organizations have done this in the United States and overseas through the Armed Forces Radio and Television networks).

Many of the agencies' information programs include fliers, fact sheets, and posters for display in prominent places. The quality of this material is rather uneven. Some of it consists of straightforward information presented without any obvious attempt to make it eye-catching or attractive. Some agencies have invested in polished, professional-looking materials with themes that are carried through in pamphlets, on posters, and in videos. At the Department of Justice, for example, the theme is "Sexual harassment in the workplace—it's against the law." Department of Agriculture materials display the theme "It could cost you your job ... and a whole lot more." A number of agencies have used inhouse resources to develop informational material; others supplement their own guidance with material written and produced by private firms. The material is generally plentiful and accessible to employees.

The objective measures available to us suggest that overall, Federal agencies' information programs have been successful in raising the workforce's sensitivity about sexual harassment and making Federal employees aware of policies, programs, and remedies. Some 92 percent of survey respondents are aware of their agencies' policies prohibiting sexual harassment, and nearly 78 percent know the formal complaint channels for victims of harassment. Further, 86 percent of respondents believe their agencies are making efforts to reduce or prevent sexual harassment. And, judging by the intensified awareness discussed earlier in this report, the information programs undertaken by Federal agencies have been successful.

Sexual Harassment Training.

All Federal agencies provide training in preventing, recognizing, and handling sexual harassment. As with their other informational programs, agencies' training varies widely. The administration of training in most departments and agencies is decentralized: subordinate organizations run their own programs, designing their own training, contracting with private sector trainers, or using training modules developed by their headquarters. At the same time, however, almost all Federal agency headquarters have set policies or minimum standards for their subordinate activities governing training content, frequency, and target audiences.

About a third of the agencies we queried mandate sexual harassment training for all employees. Almost all the rest require the training for executives, managers, and supervisors, but even in those agencies, training is very widespread among nonsupervisory personnel as well. Some 87 percent of the supervisors who responded to our survey and 77 percent of nonsupervisors said they had received sexual harassment awareness training. Some 88 percent of respondents who had received sexual harassment training had attended the training within the 2-year period preceding the survey.

One would think, given the considerable amount of sexual harassment training that Federal employees have access to or are required to attend, that training would be a significant factor in preventing harassment or in changing the way employees relate to one another in the workplace.

Sexual harassment training is, in fact, viewed in a positive light by survey respondents, but not overwhelmingly so. Nearly 76 percent of survey respondents identified training for all employees as one of the most effective actions an organization can take to reduce or prevent sexual harassment. At the same time, while 63 percent said they think training helps prevent sexual harassment to a moderate or great extent, nearly one in five respondents indicated that training does little or nothing to help.

Of those who have attended sexual harassment training, about 65 percent said it made them more sensitive to the issues or more aware of the feelings of others. On the other hand, more than one in five employees who have received training said it didn't really affect their attitudes or beliefs. Of course, some in this group of respondents may already have been sensitive to the issues prior to attending formal training. However, more than 1 in 10 said training had made them "more skeptical about issues surrounding sexual harassment." It appears that for all the success that the training may have achieved in sensitizing the workforce, there are still improvements to be made, as is evident from these reactions of respondents, as well as the fact that the amount of sexual harassment reported by respondents has not declined over the years.

Training evaluation. If some employees view training as ineffective, if sexual harassment persists while training to prevent it is nearly universal, is something lacking in the training? To answer this question agencies must devise ways to evaluate training that go far beyond asking trainees how they liked the class.

When we asked Federal agencies whether they had conducted any evaluations of the content, effectiveness, or quality of the sexual harassment training they offer employees, the responses were similar from agency to agency. Most of the evaluations consist of participant critiques completed at the end of the training. Course evaluations such as these are very common, but are of limited use. As noted in the Board's recent report on training in the Federal Government, "[W]hile this sort of evaluation probably says something about how participants felt immediately after having received the training, it most likely says very little about what the participants took back to the job."12

None of the agencies described evaluations that examined whether the training had any effect on particular problems that had been identified in their organizations. Without more meaningful evaluations, agencies won't really know how much difference training makes, and what content changes might further curtail sexual harassment. The investment agencies make in training—in terms of employee time in class as well as dollars to buy or develop the courses—is significant. For this reason, and to ensure that the right topics are chosen and adequately covered, agencies should evaluate their sexual harassment training programs in more depth than they currently do.

Training content. When we asked agencies what they covered in their sexual harassment training programs, they all listed the full range of issues surrounding sexual harassment, including legal issues, definitions, roles and responsibilities of employees and management, confidentiality, reprisals, gender awareness, and assertiveness. Survey results and written comments from a number of respondents suggest that this last topic may deserve increased emphasis. As noted previously in this report, assertive responses to sexual harassment have proven to be the most effective, but many employees are unwilling or unable to use them. Training is among the remedies recommended by some of our respondents:

Employees should perhaps take courses or read materials telling them how to say no. Until that becomes the normal response to unwanted advances, there will always be ambiguity and misunderstanding.

If I am unable to handle the situation on my own, I have no problem taking it to someone who can. I am concerned for persons who cannot confront this issue and I feel programs, training, and surveys are much needed.

While it should be stressed that Federal agencies all told us that assertiveness is a topic included in their sexual harassment awareness training, none of them described programs that make assertiveness a dominant element in the training program. Emphasizing this topic in the training (as well as in other aspects of their awareness programs) is one way in which Federal agencies can help individuals facing harassment to communicate their wishes clearly to the perpetrator, an outcome that benefits both the employees and the agencies where they work.

Handling Complaints.

Consistent with what we learned from survey data (that only 6 percent of victims report having taken formal action in response to harassment), Federal agencies report that they receive a relatively low number of formal complaints. During the entire period covered by fiscal years 1991, 1992, and 1993, a total of 1,435 formal sexual harassment complaints (an average of 478 per year) were filed in 20 of the Federal departments and agencies that provided us with information for this study.

Nonetheless, this total represents a major increase in the number of formal complaints since the 1980's, notwithstanding a decrease in the size of the workforce.13 At that time the major departments and agencies reported a total of 1,008 formal complaints alleging sexual harassment during the 7-year period FY 1980 through FY 1986 (about 144 per year). This increase, like the increase in incidence rates, may at least in part reflect the intensified awareness of sexual harassment issues.

It's probably no accident that most victims, when they took action at all, handled the situations informally. In addition to employees' reluctance to take formal action, agencies work actively at encouraging informal resolution to workplace problems associated with unwanted sexual attention. All agencies have the usual formal channels for redress; i.e., procedures for filing grievances and discrimination complaints. In most agencies there are also grievance procedures developed under collective bargaining agreements. But beyond the more formal means of dealing with harassment, is a wide variety of organizations and persons whom agencies make available to assist victims, including the Federal Women's Program staff and counselors, personnel office staff, EEO counselors, and employee assistance program counselors.

In addition, several agencies have set up sexual harassment hotlines that allow employees to seek assistance but remain anonymous if they wish. And a few agencies told us that in handling complaints they encourage employees who experience harassment first to confront the harasser (either in person or in writing), and then report the situation to the supervisor or the harasser's supervisor. A great deal of well-publicized help of all kinds currently is available for victims, so it isn't surprising to find both more formal complaints being filed, and more informal actions undertaken to stop harassment.

Additional Measures Agencies Need to Consider.

In addition to policy, training, and complaints programs (the major vehicles agencies use to respond to sexual harassment), agencies need to consider several more tailored responses that deserve particular attention. These are responses that can help agencies better identify the nature and extent of sexual harassment within their organizations, and better assist victims when regular programs aren't enough. The additional efforts and the issues they address are discussed below.

Agency-specific assessments. Because diagnosing the nature, source, and size of a problem is an important step in arriving at a solution, we wanted to know whether Federal departments and agencies had conducted surveys to pinpoint sexual harassment issues specific to their own organizations. What the agencies told us about their internal studies and assessments suggests that agencies have not been aggressive enough in this area.

Nine of the twenty-two largest departments and agencies reported that they had not conducted sexual harassment surveys of their employees. In 4 of the 22 agencies data on sexual harassment are occasionally collected in the process of interviews conducted for other purposes, such as EEO, Inspector General, or personnel management reviews.

In six other agencies sexual harassment surveys and studies have been conducted in some—not all—of the subordinate organizations. The Department of Energy reported that it collects data on its workforce's experience with sexual harassment through several items on the evaluation form that employees complete following sexual harassment training. The Department of State has conducted two agencywide sexual harassment surveys, and the Department of Justice reported that it plans to survey its workforce in the near future.

Agencies need to ensure that the content and goals of their programs to overcome sexual harassment in the workplace are directly linked to what is known about the nature and extent of the problem. Studies and surveys can help an organization's policymakers see problems through employees' eyes. They can also help devise targeted remedies, sensitive to an organization's own special culture, to address the organization's own special problems. Knowing what and where the most serious problems are should help agencies concentrate scarce energy and resources in the most efficient way.

Victim assistance. For all of the efforts made by Federal agencies to establish effective policies, administer useful programs, and encourage victims of harassment to seek the appropriate assistance, sometimes the "recommended approach" just isn't enough to stop harassment. There are employees who are unable to confront their harassers, and harassers who, when challenged, treat the situation as a joke. Sometimes the supervisor is the harasser. Sometimes second-line supervisors are no help because they react to criticism or complaints by closing ranks. Counselors may not always be able to gain the trust of victims who are worried about breaches of confidentiality. And sometimes victims fear that reporting inappropriate behavior will have an impact on the harasser much harsher than they intend—they just want the behavior to stop and are not particularly interested in exacting punishment.

A number of survey respondents described their concerns about where a victim can go for help when the usual approaches are inadequate or unsuccessful. Although all Federal departments and agencies provide counselors of some kind (typically EEO, personnel, or employee assistance program officials, or other employees with EEO collateral duties), these sources of assistance are sometimes seen as lacking either the independence or the authority to really make a difference.

One agency that provided information for this study described a new program that accommodates some of these employee concerns. The Department of Justice has established a program, outside the existing EEO and grievance processes, that requires its bureaus to appoint collateral duty contact persons who are available to advise alleged victims of harassment. The contact persons listen objectively, inform the complainants of their options, and ensure a quick, impartial, and discreet inquiry into the situation. Each contact person has a link with a designated management official to whom the more serious issues can be raised.

The contact persons are located at levels in their organizations that make them accessible to alleged victims of harassment. They afford the alleged victims the opportunity to deal with their problems in a relatively fast and informal way. This approach makes management more receptive to resolving problems because the problems can remain inhouse and not be elevated to higher levels within the department. The program is too new for results to be available, but the Department of Justice has indicated it plans soon to evaluate the program's first year of operation.

This program has elements similar to what some survey respondents indicated they would find useful in responding to sexual harassment—a kind of ombudsman, or network of such individuals, who have the authority to investigate complaints and who can help achieve a solution. Ideally, these advisors would be people to whom complainants could go without fear of reprisal and with assurances that someone would listen to their story, take it seriously, keep it confidential, and do something to help.

CHAPTER 6

Court Decisions and Evolving Views of Sexual Harassment

To understand sexual harassment in context it is important to be aware of the course the law and the courts have taken in recognizing, defining, and correcting sexual harassment. Since the last Board study of sexual harassment, important judicial decisions, including a Supreme Court ruling in 1993, have continued to change the way courts view sexual harassment.

We've arrived at the current understanding of sexual harassment as a legal concept in a step-by-step fashion since the Civil Rights Act of 1964, in title VII, first made various kinds of workplace discrimination illegal. At first the courts did not view sexual harassment as a cause for legal action. The law, after all, had not explicitly mentioned sexual harassment as an illegal act. Thus, in a case decided in 1975, a woman who alleged that her job had been abolished because she refused sexual relations with her boss was found not to have been discriminated against. The supervisor's actions, the court held, were based not on her sex, but on her rejection of his advances.14 Similarly, a 1975 decision held that two women who quit their jobs rather than continue to be subjected to their supervisor's sexual advances had not suffered sex discrimination. In this case, the court reasoned that the supervisor was following a "personal urge," not a company policy, and furthermore, if such behavior were cause for legal action under Title VII, every time an employee made an amorous advance there would be a potential Federal lawsuit.15

But these judicial rulings were not destined to stand. By the late 1970's both cases had been appealed and the decisions reversed. The appellate courts found that Title VII did apply in these cases, that sexual harassment was, indeed, a form of sex discrimination, and was therefore against the law.16

EEOC Guidance

Other cases brought to court during those years continued to refine the legal relationship between sexual harassment and sex discrimination, and in 1980 the U.S. Equal Employment Opportunity Commission issued guidelines that reflected the growing body of judicial decisions that interpreted sexual harassment as a form of discrimination on the basis of sex. The EEOC guidelines, which continue in effect today, explicitly state that sexual harassment violates the law that forbids sex discrimination (Title VII of the 1964 Civil Rights Act). The guidelines also lay out the kinds of behavior and conditions that must be present for actions to constitute sexual harassment.17

It is the employer who may be held responsible, even though other employees or nonemployees may be the source of the harassment.

According to the guidelines, unwelcome verbal or physical conduct of a sexual nature is sexual harassment when:

1. An individual's rejection of such conduct—or submission to it—is used as a basis for employment decisions that affect the employee; or
2. The unwelcome conduct interferes with an employee's work performance or creates an intimidating, hostile, or offensive working environment.

The EEOC guidelines and subsequent court decisions have led to the commonly held concept that there are two principal forms of sexual harassment, quid pro quo harassment ("this for that"), and hostile environment harassment. In quid pro quo harassment, an employee is expected to submit to some sort of sexual demand(s) in exchange for continuing employment or an employment-related benefit such as a promotion, good performance rating, or award nomination. This type of harassment, which typically involves a harasser who has authority over the victim, is clearly an abuse of power.

Hostile environment harassment often is less obvious than quid pro quo harassment, but it can be equally serious with equally adverse consequences. This kind of harassment involves unwanted behavior of a sexual nature that creates a hostile or offensive work atmosphere that may interfere with an employee's job performance. It does not require that the employee have experienced actual economic loss, and it is not necessarily a supervisor or person in authority who creates the hostile environment. Coworkers are just as likely as superiors to be the source of hostile environment harassment, and nonemployees, too, may be guilty of creating the offensive environment. It's important to note, however, that it is the employer who may be held responsible, even though other employees or nonemployees may be the source of the harassment.

Supreme Court's First Sexual Harassment Decision

Throughout the early 1980's judicial rulings were being issued that found illegal discrimination in cases involving unwelcome sexual behavior, both those in which quid pro quo harassment had occurred and those that involved a hostile work environment. Then, in 1986, in its first decision on sexual harassment (Meritor Savings Bank, FSP v. Vinson), the Supreme Court addressed a number of critical issues, including quid pro quo and hostile environment harassment. In this case, a woman who had engaged in sexual relations with her boss was found to have been illegally discriminated against because, although her actions were voluntary, his advances were unwanted and she feared reprisal. The Court's decision, which endorsed the EEOC guidelines, established a number of important standards, including the following:

• A hostile or abusive work environment created by discrimination based on sex can be a violation of Title VII of the Civil Rights Act;
• To be a cause for legal action the harassment must be severe enough to alter the conditions of the victim's employment (remarks that are offensive but not pervasive usually won't qualify);
• A sexual harassment claim does not have to involve a negative economic effect on the victim;
• The crucial point in such cases is whether or not the harasser's sexual advances were unwelcome rather than whether or not the victim submitted voluntarily to those advances;
• Employers are not protected from liability merely because their organizations have grievance procedures and antidiscrimination policies; and
• Employers may be held liable for sexual harassment committed by their supervisory employees, sometimes even in cases in which the employers were not aware of the behavior, but employer liability for what supervisors do is not automatic.18

Although sexual harassment as a form of sex discrimination is now an accepted concept, in the years since this Supreme Court ruling, U.S. courts have continued to address definitions of important aspects of sexual harassment that were left unclear in the Meritor decision. For example, in judging whether conduct is severe enough to justify legal action, many courts applied a "reasonable person" concept; that is, considering whether a hypothetical reasonable person would have found the alleged harassing behavior offensive. And since that time, the concept has further evolved. Some courts have asserted that a "reasonable woman" standard should be applied because men and women often view the same behavior in very different ways.

A case decided by the circuit court of appeals in California in 1991 (Ellison v. Brady), involved a male employee who desired a relationship with a coworker and sent her unsolicited notes about his feelings. The court had to decide whether his conduct had altered the conditions of the coworker's employment by subjecting her to an abusive working environment. In making its determination this court applied the reasonable woman standard (rather than a reasonable person standard), noting that in evaluating the seriousness of harassment "we should focus on the perspective of the victim."

Supreme Court's Second Sexual Harassment Decision

When, in 1993, sexual harassment issues reached the Supreme Court for the second time, the court did not discuss the reasonable woman standard in its decision. But it did support unanimously the view that Title VII of the Civil Rights Act of 1964 does not require the victim to have suffered concrete psychological harm before the harasser's behavior can be considered an unlawful employment practice.

The case, Harris v. Forklift Systems, Inc., involved a woman whose supervisor frequently addressed her in an offensive and derogatory way. Lower courts had found this behavior vulgar and demeaning, but not a matter of illegal discrimination because it had not severely affected the woman's psychological well-being. The Supreme Court saw the situation differently. Even without causing psychological harm, the Court maintained, an abusive atmosphere can detract from employees' job performance, discourage them from staying on the job, or keep them from advancing in their careers.20

In deciding for the victim, the Court reiterated the standard it had applied in its Meritor decision 7 years earlier—if a workplace is permeated with ridicule and insult that is sufficiently severe to alter the conditions of the victim's employment, the law (Title VII) has been violated. The Court said that in applying this standard it was taking a "middle path" between making just any offensive conduct a cause for legal action and requiring a tangible psychological injury.

Offensive words or behavior should be stopped before the provisions of Title VII have been breached.

This "middle path" adopted by the Court reflects the true complexity of the issues. Although employers and employees both may yearn for an air­tight definition of legally actionable sexual harassment, such a definition is not very likely as long as human relationships remain complicated and human behavior continues to be interpreted in an infinite number of ways. Simply put, sometimes offensive conduct is illegal and sometimes it's not. And the courts, to date, have provided no precise formula to indicate when each standard applies.

Although the Supreme Court's second decision on sexual harassment did not describe specifically the behaviors that create an environment that would be cause for legal action, it did describe the factors to consider in reaching conclusions about whether conduct is legally actionable. The Court said that all the circumstances need to be considered, including

• How often the conduct occurs;
• How serious the conduct is;
• Whether the behavior physically threatens the victim, or stops at offensive comments; and
• Whether the behavior unreasonably interferes with work performance.

The Court also noted that the victim must perceive the environment to be abusive in order for the conduct to be considered illegal. If the victim doesn't perceive the environment as hostile or abusive, Title VII has not been violated.

Given the evolution of legal definitions and judicial decisions about sexual harassment, employers today possess a degree of certainty about some issues related to sexual harassment (e.g., sexual harassment can equate to illegal discrimination), yet face a continuing level of ambiguity about other issues (e.g., when does offensive behavior cross the line into illegal sexual harassment?). Thus, judging situations on a case-by-case basis, considering all the circumstances involved, and attending to the view of matters through both the eyes of the beholder (i.e., the victim) and the reasonable person (or woman) are accepted approaches in dealing with sexual harassment cases.

In the final analysis, however, while Federal managers must of course be concerned about the occurrence of illegal acts of sexual harassment in the workplace, their more fundamental concern should be what kind of workplace environment is provided for Federal workers. It should not be the goal of Federal managers merely to eradicate the kinds of behavior that cross the line from offensive to illegal. Offensive words or behavior in the workplace should be stopped before the situation reaches the point where the provisions of Title VII have clearly been breached. Any behavior that is vulgar or disrespectful or insensitive must be an abiding concern for managers and supervisors.

CHAPTER 7

Summary, Conclusions, and Recommendations

After all that Federal workers and observers of the Federal sector have said, done, written, and thought about eliminating sexual harassment from the workplace, the problem persists. At the same time, progress has been made in building an awareness of sexual harassment, an understanding of the relevant issues, and sensitivity to the way people treat other people at work.

While the Federal Government has done a great deal to address sexual harassment in the workplace, refining and refocusing of programs and policies are necessary to continue this progress. Managers and supervisors need to make it clear that they care how their employees treat one another. Agencies need to identify their worst problems and best programs and tailor their future efforts accordingly. As the workforce is reduced and agency budgets decrease, there is no corner of the Government wherein the Nation can afford to tolerate conduct that diminishes productivity and erodes morale.

Conclusions

Undoubtedly, progress has been made in educating the workforce and raising awareness about sexual harassment since our first study of sexual harassment in 1980. Over 87 percent of Federal supervisors and 77 percent of nonsupervisory employees have now received training related to this problem. Some 78 percent of employees said that they know the channels to follow if they have been harassed and want to report it. All agencies have policies prohibiting sexual harassment, and 92 percent of Federal employees are aware of those policies.

Even the fact that the sexual harassment incidence rate remains high can, in a certain sense, be construed as a type of progress. For example, the amount of harassment reported could be viewed as an indicator that the workforce has been educated as to what may constitute sexual harassment and is now categorizing those behaviors as such. This doesn't mean that the Government has created a workforce of complainers; it means, rather, that employees are learning that it's appropriate to call attention to situations that interfere with their work.

Federal employees are defining sexual harassment more broadly than in the past, and more of them are defining more forms of unwanted sexual attention as sexual harassment. At the same time, there is a fairly widespread desire among employees for more precise definitions of sexual harassment, something like a list of do's and don'ts that would take the guesswork out of whether one's conduct or comments are likely to be construed as sexual harassment.

In a sense, the absence of such definitions has had a positive effect on the workforce, as men and women have been moved (willingly or unwillingly) to think more critically about what they say and do. A heightened sensitivity about what behaviors may or may not be considered sexual harassment has stimulated concern among Federal employees about how others will perceive their words and actions. Furthermore, there is no strong evidence that the remaining uncertainties have caused a widespread chilling effect in the workplace, as some had feared. While many survey respondents worry about how their conduct might be interpreted, most did not agree that fear of being accused of sexual harassment had made their organizations uncomfortable places to work.

Federal agencies believe that, overall, the spotlight on sexual harassment has improved the situation in the workplace. They believe that while there have been increases in both formal and informal reports of sexually harassing behaviors, this stems not from an increase in the amount of harassment but from the fact that the workforce is better educated about it and more people feel comfortable coming forward to report it.

Greater awareness of the issues surrounding sexual harassment, more people willing to report unwelcome sexual behavior, and more overall sensitivity to the feelings and perceptions of others may be among the factors contributing to a general decrease in the ill effects of sexual harassment in the Federal workplace. Since the Board's last sexual harassment study, there has been a significant drop in turnover and sick leave used in response to sexual harassment as well as a decline in the severity and duration of productivity losses resulting from the disruptive effects of sexual harassment.

Nevertheless, the sexual harassment incidence rate of 44 percent for women and 19 percent for men for the 2 years preceding our survey is evidence that much more progress must be made in consistently implementing sexual harassment policies, following through with appropriate remedies or penalties, and improving training. Maintaining—and expanding—employee awareness is essential. Equally important is the fact that sexual harassment program activities and the resources to support them should be focused on identified problem areas rather than spread to an equal degree to every part of every Federal organization.

Our study provides a very broad perspective on sexual harassment as it now exists in the Federal workplace. It remains up to departments and agencies to pinpoint the worst problems and adapt programs to enable managers and supervisors to deal with them expeditiously.

Recommendations

1. Agencies should find ways to capitalize on what is already known about the most effective actions that can be taken to prevent and eliminate sexual harassment; that is, they should publicize penalties and encourage assertive actions on the part of employees who are targets of unwanted sexual attention.

The results of all three MSPB sexual harassment surveys have shown that employees believe that publicizing sexual harassment policies and penalties are among the most effective actions agencies can take to prevent sexual harassment. The nearly universal awareness of sexual harassment policies among members of the Federal workforce indicates that agencies have done a good job in getting the word out about their policies. Less is known among the workforce about what happens to people who harass others. Employees should be made aware of how the agency intends to discipline proven harassers. Victims should always be informed about what happened to their harassers, and penalties should be public enough to serve as examples to potential harassers that management's prohibition of sexual harassment is more than lip service.

As indicated earlier, the most effective approach for targets of unwanted sexual attention is to take assertive actions such as confronting harassers and telling them to stop or reporting the behavior to someone in a position to help. Agencies should facilitate this approach by highlighting assertiveness in their training programs and by making it easier for victims to report harassing behaviors through informal programs such as neutral advisors or an ombudsman who serves as a confidential consultant to victims.

2. Managers and supervisors should be firm and consistent in penalizing proven harassers.

When harassment occurs, managers and supervisors should take action based on the seriousness of the offense rather than the rank of the offender. In deciding a reasonable penalty to be imposed when harassment has been proven, managers and supervisors should not give undue weight to the harasser's performance and value to the agency. Managers and supervisors must understand that the value of a harasser's contributions to the organization is likely to be diminished by behavior that hurts morale, demonstrates a lack of ethics, or exhibits a double standard. Further, the example that management sets in following through with appropriate penalties can be more effective as a preventive measure than the policies it promulgates.

3. Agencies should diagnose the extent and seriousness of sexual harassment within their own organizations so that they know what kinds of solutions are appropriate and where resources should be concentrated.

The content and goals of agency programs to eliminate sexual harassment should be linked directly to what is known about the nature of sexual harassment in the agency. Studies and surveys that help agency policymakers see the work environment through employees' eyes can help in devising remedies that are sensitive to the agency's multiple cultures; e.g., headquarters activities, field activities, administrative operations, health care facilities, law enforcement operations, scientific laboratories. Knowing what and where the most serious problems are can help agencies target scarce energy and resources in the most efficient ways. As much as sexual harassment costs the taxpayer, and as lean as future agency budgets are likely to be, Federal organizations cannot afford to direct insufficient attention to serious problems while expending resources in areas where problems are minimal or nonexistent.

4. Agencies should evaluate the effectiveness of the sexual harassment training they provide to ensure it addresses identified problems. Agencies should pay particular attention in their training efforts to the problem of sexual harassment by coworkers.

Sexual harassment training is provided in every agency at all organizational levels. This training represents a considerable investment, and while most agencies know whether or not it's popular with participants, they generally don't know what kind works best, what parts of it are effective, what kinds make no difference, and whether any of it has a negative effect. Therefore, agencies should adapt their training to address what they learn from their own self-diagnoses of the extent of sexual harassment in their workforces and from studies such as this one. For example, because findings consistently show that coworkers and other employees are the primary source of sexual harassment in Federal agencies, training efforts might emphasize strategies for handling harassment from peers. Followups should be conducted to determine what effect, if any, training actually has on the targeted workforce, and training content should be revised if it is found to make no appreciable difference in preventing or stopping sexual harassment. Agencies must also ensure that the emphasis given to important programs such as sexual harassment training be given the proper type and degree of emphasis, in keeping with what is known about the nature and extent of the problem in their own organizations.

Appendix 2

MSPB Sexual Harassment Survey
Percentage of male and female respondents who said they had experienced the indicated unwanted behaviors in the previous 2 years

Appendix 3

MSPB 1994 Sexual Harassment Survey
Percentage of male and female respondents who said they had experienced the indicated unwanted behaviors once and more than once.

Note: Not shown are percentages of respondents who indicated they had not experienced any of the indicated behaviors. Numbers are rounded

Appendix 4

MSPB 1994 Sexual Harassment Survey
Percentage of male and female respondents with the indicated characteristics, by victim and nonvictim status

Appendix 5

MSPB 1994 Sexual Harassment Survey
Percentages of male and female respondents who said they had taken the indicated action in response to unwanted sexual attention

Appendix 6

MSPB 1994 Sexual Harassment Survey
Percentage of victims who said that the indicated formal action made things better, made things worse, or made no difference

Endnotes:

* Notice: This U.S. Merit Systems Protection Board report was electronically converted from its desktop publishing format into an accessible format. Because of this conversion, there will be some slight variations (especially in special characters) from the paper copy. Notably, page numbers in the body of the report vary by eleven pages from those shown in the Table of Contents, and Appendix 1 could not be included in this electronic version.

For a paper copy of the report, please contact: U.S. Merit Systems Protection Board, Office of Policy and Evaluation, 1120 Vermont Avenue, NW, Room 884 Washington, DC 20419 -- Toll-free (800) 209-8960, V/TDD (202) 653-8896, FAX (202) 653-6772, Internet: pe@mspb.gov.

1 The employees who participated in the survey were selected randomly using the Central Personnel Data File maintained by the U.S. Office of Personnel Management. This file is a computerized data base with information on about 2 million Federal civilian employees. Not included in the data base are employees of the U.S Postal Service and other agencies not required to report personnel statistics to OPM, such as the U.S. Central Intelligence Agency.

2 We drew our employee sample from these agencies: the Departments of Agriculture, the Air Force, the Army, Commerce, Defense, Education, Energy, Health and Human Services, Housing and Urban Development, the Interior, Justice, Labor, the Navy, State, Transportation, the Treasury, and Veterans Affairs, as well as the Environmental Protection Agency, the General Services Administration, the National Aeronautics and Space Administration, the Office of Personnel Management, the Small Business Administration, and a cross section of other, smaller agencies.

3.Harris v. Forklift Systems, Inc., 114 S.Ct. 367 (1993)

4.To facilitate comparison with the data from the Board's previous sexual harassment studies, we use incidence rates for the 1994 survey discussion that do not take into account people who reported having experienced stalking. However, we found that when victims of stalking were factored in, the overall incidence rates were identical to the rates without stalking (44 percent for women, 19 per­cent for men). This was the result of the fact that stalking is rare (only 4 percent reported it) and when stalking occurs, usually one or more of the other harassing behaviors also are present, so survey respondents who reported the behavior are counted as victims either way.

5. The average adjusted basic pay rates, which include base pay and locality pay, but not the cost of benefits, are derived from data from OPM's Civilian Personnel Data File, September 1993. At that time the average annual basic pay was $42,066, for men and $31,931 for women.

6. U.S. Equal Employment Opportunity Commission, "Federal Sector Report on EEO Complaints and Appeals by Federal Agencies for Fiscal Year 1993," Washington, DC, September 1995, p. 29.

7.Adverse action appeals can be prompted by sexual harassment under a number of circumstances. For example, an employee might appeal a demotion if she believed the reason for the action was related to her refusal to accept a date with a supervisor.

8. Stephanie Riger, "Gender Dilemmas in Sexual Harassment Policies and Procedures," American Psychologist, May 1991, vol. 46, No. 5, pp. 497-505.

9. Dennis K. Reischl and Ralph R. Smith, The Federal Manager's Guide to Preventing Sexual Harassment, FPMI, Inc., Huntsville, AL, 1989, p. 18.

10. U.S. Merit System Protection Board, "First-Line Supervisory Selection in the Federal Government," Washington, DC, June 1989, p. 4.

11. Human Resources Executive, December 1993, vol. 7, No. 13, p. 39.

12. U.S. Merit Systems Protection Board, "Leadership for Change: Human Resource Development in the Federal Government," Washington , DC , July 1995, p. 20.

13. The population represented in the Board's 1987 sexual harassment study totaled about 2 million; the 1994 population was about 1.7 million.

14. Barnes v. Train, 13 Fair Empl. Prac. Cas. (BNA) 123 (D.D.C. Aug. 9, 1974).

15. Come v. Bausch and Lomb, Inc., 390 F.Supp. 161 (D.Ariz. 1975).

16. Barnes v. Costle, 561 F.2d. 983 (D.C. Cir. 1977); and Come v. Bausch and Lomb, Inc., 562 F.2d. 55 (9th Cir. 1977).

17. The EEOC guidelines are found at 29 Code of Federal Regulations 1604.11.

18. Meritor Savings Bank, FSB v. Vinson, 106 S.Ct. 2399 (1986).

19. Ellison v. Brady, 924 F.2d 872 (9th Cir. 1991).

20. Harris v. Forklift Systems, Inc., 114 S.Ct. 367 (1993).