The U.S. Equal Employment Opportunity Commission

II. Literature Review

The literature on alternate dispute resolution methods, including mediation, can be divided into two categories: theoretical and empirical. Most of the theoretical literature attempts to define mediation and point out its advantages and disadvantages,<9> establish the protocols for mediation,<10> and/or speculate on the uses of mediation in different contexts.<11> The empirical literature includes case studies and surveys.<12> Since mediation is a rather new field of academic research, there is not a significant body of empirical research in the field. It is not our intention to review the entire literature on mediation. We will review both theoretical and empirical literature to the extent to which they aid our investigation, which is to measure participant experiences with and opinions of the EEOC mediation program.

This literature review is organized into three subsections. The first subsection defines mediation and discusses its uses, advantages, and disadvantages. The second subsection looks at "justice and ADR program evaluation." The third subsection depicts the empirical studies that evaluate the performance of mediation programs.

A. Mediation: Background, Definitions, Uses, and Advantages

During the past few years, ADR methods have gained increased acceptance in the U.S. workplace. The report of the Commission on the Future of Worker Management Relations (The Dunlop Commission), which included an investigation of the usefulness of ADR methods in resolving workplace disputes, provided an additional impetus for the examination of the merits of ADR. The Dunlop Commission pointed out two factors that beg for alternatives to the traditional system of employment litigation. First, under the traditional litigation system low-wage workers were not faring as well as the higher paid workers since they might not have the time or money to pursue a court case. Thus, the traditional system did not serve all people equally well. Second, the traditional litigation system was dominated by ex-employees, rather than by employees who sought to redress complaints while continuing employment. This shows that for those employees who want to continue with their current employer, the traditional system did not offer much assistance. Given these patterns, the Dunlop Commission recommended that ADR methods be encouraged.<13>

From the Grecian city states to the 21st century USA, mediation has brought together parties to a dispute in search of solutions. Mediation has been found to be an effective dispute resolution tool for producing settlements to disputes, while providing a high degree of disputant satisfaction with the process.<14> This use of mediation to resolve employment law disputes is receiving increased attention from attorneys and scholars. For example, on July 8, 2000, the American Bar Association, during its annual convention, conducted a program co-sponsored by its Section of Dispute Resolution and the Section of Labor and Employment Law entitled "The Mediation of Discrimination Complaints: Serving the Parties' Interests or Undermining Public Policy."<15>

The EEOC defines mediation as an informal process in which a neutral party assists the opposing parties to reach a voluntary, negotiated resolution of the charge.<16> It "gives the parties the opportunity to discuss the issues raised in the charge, clear up any misunderstandings, determine the underlying interests or concerns, find areas of agreement, and ultimately, to incorporate these ideas of agreements into resolutions."<17> The actors in the mediation are the parties themselves, assisted by a trained mediator. Thus, mediation attempts to focus on the relationship of the parties, rather than on legalism.<18> This becomes important if parties want to preserve their relationship, which may be destroyed or damaged by the adversarial nature of other dispute resolution mechanisms, such as litigation. Experts opine that ADR methods, such as mediation, become very important in non-discharge employment cases, where the parties want to preserve their relationship with their employer.<19>

Mediation has been found effective in dealing with a number of different types of disputes. It is effective not only in settling small claims,<20> but also resolving a variety of different disputes including labor grievances,<21> contractual terms, construction damages, personal injuries, damages to property and the environment,<22> and employment discrimination under the Title VII of the Civil Rights Act.<23> The advantages of mediation are many: it is less costly, it is less damaging to the relationship of the concerned parties, and it promotes a win-win solution.

According to the EEOC, the advantages of mediation are:

The key words used by the EEOC in describing the advantages of mediation are efficiency, fairness, confidentiality, non-admission of violation of laws, and speediness of resolution. Critics point out that mediation and other ADR methods "amount to a private justice system that does not always protect the public's interest in procedural fairness or disclosure of how disputes were resolved."<25>

The EEOC has addressed the public policy issue by use of a charge classification system. Charges that are identified as "A" charges usually are not selected for mediation. These charges involve cases where a reasonable case finding is highly likely or where important pattern or practice/systemic issues or other public policy concerns militate against the use of pre-investigation mediation. Where a party requests that an "A" charge be mediated, the District Director and Regional Director have the discretion to allow such mediation. However, this is the exception and not the rule for such cases.

Charges classified as "B" charges are charges where further investigation is required to make a determination concerning their merit. In general "B" cases are eligible for pre-investigation mediation. However, "B" cases that involve the Equal Pay Act or pattern or practice/systemic allegations are not eligible for pre-investigation mediation.

Intake personnel usually make the initial classification decision. Each office has a process that reviews and ensures the accuracy of its classification decisions. It should be noted that the Equal Pay Act cases, whether they are classified as "A" or "B," are not eligible for pre-investigation mediation.

B. Justice and Dispute Resolution Program Evaluation

How does one evaluate the effectiveness of a dispute resolution program? Participant feedback is one of the essential criteria used to assess a system's effectiveness and to suggest improvements to the system. Any evaluation of a dispute resolution method should include participant feedback regarding their perceptions of procedural justice (justice with regard to the process) and distributive justice (justice with regard to the outcome).

1. Procedural Justice

Procedural justice can be defined as "the perceived fairness of the process through which decisions are made."<26> The theory of procedural justice posits "disputants prefer procedures that provide them with voice, control over the outcome, and fair treatment by the third party."<27> Thus, parties to a dispute must first be given a fair chance to voice their concerns. Second, parties must have control over the outcome of mediation since mediation is about self-determination. Third, the mediator must be perceived as, and be, fair and neutral. Measurement of participant perception of procedural justice regarding mediation must consider whether the participants felt that they were treated with fairness by the mediator, their voice was heard, and they had control over the outcome.

Fairness is an essential element of procedural justice. It is especially crucial to any mediation program because it is a necessary ingredient to induce both participation and settlement.<28> Hodges notes that the EEOC has recognized not only the importance of actual and perceived fairness in its mediation program, but also that "fairness requires adequate information, the opportunity for assistance, knowing and voluntary participation, neutrality, confidentiality, and enforceability." The EEOC ADR programs are developed according to these principles of fairness.<29>

As indicated above, an element of procedural justice is "knowing participation," i.e., participant understanding of the process. Research in organizational theory has shown that understanding is an important factor in employee attitudes towards organizational activities.<30> Hence, understanding of the process is essential for participant satisfaction with the process and for their perception that the process was fair. Hodges states "in order to make an informed choice about whether to participate in mediation, the potential parties need information about the mediation process, statutory rights and remedies, and the advantages and disadvantages of both mediation and litigation." She also points out that both the EEOC and Administrative Conference of the United States (ACUS) recognize that ADR programs must be fair to the participants, and fairness requires that extensive information be provided to the participants about the process. It should be noted that one of the recommendations of the Task Force on Alternative Dispute Resolution in Employment, which established the "Due Process Protocol for Mediation and Arbitration of Statutory Disputes Arising Out of the Employment Relationship," is that participants be "given access to information about both the process and the substance of the dispute."<31>

The "timing of mediation" and "representation" are two other important variables that affect procedural justice. Prompt scheduling of a program is considered to be an indication of effective program management. The timing of mediation is important since one of the touted advantages of mediation is that it is less time consuming than other methods of dispute resolution. Once the case is referred to mediation and if it takes place promptly (i.e., before positions harden), a settlement may be more likely."<32>

Fairness/procedural justice requires that there is opportunity for assistance to the participants. Representation in the form of either attorneys or other knowledgeable persons, for one or both parties, may serve to balance power.<33> It is important not only to let the participants know that they can bring in representatives, but also to notify each party as to who will be representing the other party.<34> There is empirical evidence to support the notion that when one party to a dispute appeared with an unanticipated representative, the other party was concerned about the fairness of the process.<35>

The mediator is crucial to the process of mediative decision-making. As the third party who assists the parties to a dispute, a mediator facilitates communication with the parties, helps them to focus on the real issues of the dispute, and assists them in generating options for settlement.<36> Mediators, unlike arbitrators, do not mandate solutions to a dispute. Instead, they are facilitators who enable parties to reach their own agreement.<37> As facilitators, it is essential that the mediators understand the nature of the dispute in question and the viewpoints of the parties to the dispute. This understanding becomes critical in clarifying the issues to the parties, which is important to promote co-operative problem solving. Mediators also serve as "a reality check, alerting parties to unrealistic expectations."<38>

One of the commonly included criteria in the definitions of mediation is that the mediator be neutral or impartial.<39> Some researchers point out that "neutrality, as traditionally practiced, actually includes two potentially conflicting qualities: (1) impartiality, which refers to the mediator' ability to maintain an unbiased relationship with the disputants; and (2) equidistance, which involves the mediator temporarily becoming aligned with each party to encourage disclosure and assist the party in expressing the case.<40> Mediators must be impartial, fair, and diligent in order to foster trust between the parties and between the mediator and the parties. They must also maintain the confidentiality of the parties.<41>

Participant perception of distributive justice or justice with regard to the outcome is another important element of participant satisfaction with the program. It is not surprising that participants would perceive the outcome to be more just when they received what they wanted than when they did not. However, even when the mediation does not end in a settlement, it has the chance to further the communication between the parties since it brings the parties together.

2. Distributive Justice

Distributive justice measures include participant satisfaction with the outcome of mediation and participant perceptions regarding the fairness of the outcome.<42> Research has shown that participants not only differentiate between procedural justice and distributive justice, but also consider procedural justice to be more important in certain situations.<43>

Gordon, in his study of the issues of justice in a union-management grievance procedure, emphasized the importance of procedural justice in workplace dispute resolution.<44> He argued "research on grievances will remain incomplete lacking the perspectives of employees who are themselves subject to the grievance system."<45> He used a five-point Likert scale that ranged from "Very Satisfied" to "Very Dissatisfied" to survey employees' overall satisfaction with the grievance system. He found that the overall evaluation of the grievance system was significantly correlated with the measures of perceived procedural and distributive justice. He also found that the correlation was significantly higher for the procedural justice factors as opposed to the distributive justice factors. Thus, his research seemed to support the findings of Folger and Greenberg<46> that the "procedures followed, rather than the outcomes obtained, have the greater influence on the overall evaluation of dispute-resolution systems."<47>

C. Empirical Studies of Mediation Programs

Given the fact that mediation is relatively new in the field of employment dispute resolution in the United States, it is not surprising that there have been few published studies on the evaluation of mediation programs. Employment law mediation programs sponsored by public agencies have been the subject of only two published studies: McEwen and Kochan, Lautsch, and Bendersky. Another study on mediation that deserves attention is by Brett, Barsness, and Goldberg. These studies are discussed in detail since they provide valuable insights into how researchers have evaluated the effectiveness of different mediation programs.

McEwen analyzed the EEOC pilot mediation program using exit surveys of participants.<48> He reported on 267 completed mediations.<49> His survey focused on the following variables: the time required to complete mediation from the filing of the charge to the close of the mediation, the duration of the sessions, the outcomes associated with the mediation, the participant satisfaction with the mediation process, the participant perceptions of the fairness of the mediation process, the participant perceptions of the mediator, and whether the participants would use the program again.

McEwen found that that the average time to complete mediation was 67 days from the filing of a charge, and that the mediation sessions averaged 3.7 hours. Regarding the outcomes of mediation, he reported that 52% of the mediations concluded with settlements and over half of these settlements provided for financial payments to the charging party. Most of the participants were satisfied with the mediation process: 66% of the charging parties and 72% of respondents expressed satisfaction with the process. Similarly, 91% of the charging parties and 93% of the respondents rated the process as fair. Regarding the mediator, 95% or more of the parties said that they trusted the mediator and indicated that they had been treated with respect during the mediation. Parties also expressed an interest in using the mediation process again. Eighty-four percent of the charging parties and 83% of the respondents indicated that they would use the mediation process again, if they had a similar problem.<50>

This research included a follow-up mail survey of 125 of the 267 charging parties. Of these 125, approximately 35% responded.<51> After several months, they continued to have positive opinions about the mediation process. According to McEwen, one limitation of the study was the short time span covered by the analysis. Also, it was not possible to follow to completion the charges filed by control group members who did not participate in the mediation program.<52>

Brett, Barsness, and Goldberg studied 449 cases administered by four private sector ADR providers—the American Arbitration Association, the CPR Institute for Dispute Resolution, JAMS-Endispute, and U.S. Mediation and Arbitration Service—to measure the effectiveness of mediation.<53> They used a mail survey of participants to evaluate, among other things, the participants' satisfaction with the ADR process that they used and their perceptions of the fairness of that ADR process. More specifically, participants were asked to indicate their responses to a series of questions, about the fairness of the mediator, the fairness of the process, their control of the process, their satisfaction with the process, and where appropriate, their satisfaction with the outcome, using a five-point scale (1 = very dissatisfied and 5 = very satisfied).

The results showed that the participants were satisfied with every aspect of the mediation. Several statements about the participant satisfaction with the process and the mediator were combined into a single index and this index yielded a mean score of 4.02 (out of 5) showing that participants were satisfied with the mediation process. Similarly, the participants who resolved their cases at mediation gave a high score (mean score of 4.08 out of possible 5) to the outcome of mediation. It should be noted that this study measured the elements of both procedural and distributive justice involved in mediation.

More recently, Kochan, Lautsch, and Bendersky evaluated the ADR program of the Massachusetts Commission Against Discrimination (MCAD),<54> a state human rights agency. The MCAD mediation program, which is voluntary, sought to resolve disputes involving claimed violation of public law that protected individual civil rights.<55> The Kochan et al. study included, inter alia, the reported results of 95 cases mediated by the Commission. Of these cases, 63% were settled.<56> The study measured claimant (charging party) and respondent satisfaction with the mediation, the influence of whether the case was settled or not on the satisfaction rate, and the participants' opinions of the mediation process and the mediator's performance.<57>

Overall, the participants rated the performance of the mediator very highly. The study found that 95% of the charging parties and 100% of the respondents answered that the mediator explained the mediation process well and 95% of the charging parties and 96% of the respondents stated that the mediator treated the parties with respect. Eighty-five percent of the charging parties and 92% of the respondents stated that the mediator listened and 75% of the charging parties and 85% of the respondents stated that the mediator provided an opportunity to express opinions. Seventy percent of the charging parties and 88% of the respondents believed that the mediator understood the dispute.

The study also reported that 63% of the charging parties and 77% of the respondents stated that they would use the mediation program again. Fifty percent of the charging parties and 68% of the respondents were satisfied with the outcome of the mediation.<58> However, only 30% of the charging parties and 35% of the respondents rated the outcome of the mediation as fair.<59> Finally, the Kochan study found that respondents were more satisfied with the overall outcome of the mediation than the charging parties (50% for charging parties and 68% for respondents).<60> Where the case was settled, charging parties and respondents were more likely to report that the mediation was fair.<61> Only about 20% of the participants in these 95 cases were surveyed. Thus, the authors acknowledged that this was "an extremely low response rate" and that this was a "major weakness" of the study.<62>

It should be noted that the MCAD program is different from that at the EEOC in several ways. First, the MCAD program uses a "fee for service provision" whereby each party in the study paid a $500 administrative fee that included three hours of the neutral's time.<63> After three hours, the mediator was free to charge his/her hourly rate, which was reported to have ranged from $100 - $350 per hour. In the EEOC program, mediation is free for its participants. Second, the MCAD program is administered by the American Arbitration Association and, thus, uses mediators found on AAA panels.<64> The EEOC uses internal staff as mediators as well as external mediators from a wide variety of sources. Third, in the MCAD program, mediation is conducted only in those cases where an investigation found cause to believe that the law had been violated,<65> whereas mediation occurs in the EEOC program before there is any investigation of the charge. Finally, under the MCAD program, the parties were required to be represented by an attorney, whereas the EEOC allows the parties to represent themselves or select any other representative of choice (i.e., accountant, spouse, wife, minister, rabbi, etc.).

Are participants who choose mediation more satisfied with the way their case is handled than those who did not? Both Kochan et al. and McEwen attempted to compare the experiences of parties who had participated in mediation with those who had remained in the traditional investigation track. Both were unable to gather the data necessary to compare the experiences of this control group with those who went to mediation.

While not a control group, McEwen surveyed charging parties who did not participate in mediation, but rather participated in an EEOC investigation. His response rate for this group was 25%, which he categorized as a "non-representative sample."<66> Of this group, only 13% of the participants indicated that they were satisfied with the way their case had been handled. Where a decision was reached in the case, only 8% were satisfied with it.

Christovich and Stallworth surveyed the persons who filed suit in federal district court after the EEOC charge regarding their perceptions on receiving a fair hearing of their claim.<67> The authors surveyed employment discrimination claimants whose names appeared on the federal district court docket in the U.S. District Court for the Northern District of Illinois. Eighty-two percent of the claimants reported that the overall federal district court litigation process was unfair. Eighty percent believed that the judge did not consider their side of the case and 68% reported that they believed that the judge was biased.<68> This research also had some limitations due to the survey response rate, which was 25%. Also, the researchers could not validate the representativeness of these responses leaving unresolved the issue of non-response bias.<69> What is clear from the McEwen and Christovich/Stallworth research is that those who use processes other than mediation do not appear to fare better than those who go to mediation.

In conclusion, the literature review shows that the measurement of procedural and distributive justice is essential to the evaluation of any mediation program. Elements of procedural justice included variables such as "knowing participation," representation, performance of the mediator, fairness of the process, and opportunity to present participant views. Elements of distributive justice included participant satisfaction with the outcomes and the fairness of the outcomes.

previous | contents | next

This page was last modified on October 2, 2000.

Home Return to Home Page