Reprinted with permission from Dispute Resolution Magazine, Fall 2017 (24:1), at 26-30. © 2017 American Bar Association. All rights reserved. This information or any or portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association.
Mediators frequently make assertions—both expressly and implicitly—about which actions and approaches are “best” in terms of their effect on mediation outcomes. Because these statements are often based on assumptions, anecdotal observations, and untested beliefs, in 2013 the ABA Dispute Resolution Section created a Task Force on Research on Mediator Techniques and asked it to review the empirical research and try to determine which actions really are beneficial, which are detrimental, and which have no effect on mediation outcomes.
The Task Force members included mediators, researchers, law professors, program administrators, and other professionals with a range of experience and expertise. The work of the Task Force produced a 69-page final report, which was adopted by the ABA Section of Dispute Resolution Council in August 2017.1
This description of the Task Force’s work and summary of its analysis ends with a general conclusion: more and better empirical research on this subject is needed, and until we have it, we should be careful in making broad claims about “best” techniques.
Studies, Actions, and Outcomes Examined
The Task Force cast a wide net, reviewing studies of any non-binding process in which a third party helped disputants try to resolve any type of conflict. To be considered relevant, the studies had to contain empirical data examining the effects of one or more mediator actions or approaches on one or more mediation outcomes.2 We reviewed a final set of 47 studies. Thirty-nine studies involved only mediation, and eight studies involved another process in addition to or instead of mediation.
The Task Force organized the wide range of mediator actions and styles examined in the studies into seven categories (see Table 1) and grouped the mediation outcomes into three categories (see Table 2). Where data was available, we examined the empirical findings regarding the effects of each category of mediator actions on each set of mediation outcomes and reported the findings separately for each of these mediator action-mediation outcome pairs.
The studies covered a range of dispute types, including general civil, domestic relations, labor management, community, and others. A majority of studies involved a single mediator and court-connected mediation. But the studies varied greatly in other characteristics of the mediators and the mediation context; whether they examined specific mediator actions or looked at mediator approaches comprised of multiple actions; how those actions or approaches, as well as outcomes, were defined and measured; the data sources and research methodology used; and the robustness of the findings.3
All this variation could contribute to different findings across the studies and made “apples to apples” comparisons extremely challenging, so drawing broad conclusions was difficult.
Empirical Findings: Effects of Mediator Actions on Outcomes
The Task Force’s compilation and analysis of existing empirical research found that none of the categories of mediator actions has clear, uniform effects across the studies—that is, none consistently has negative effects, positive effects, or no effects—on any of the three sets of mediation outcomes.4 For a majority of the action-outcome pairs, an equal or larger number of studies found that mediator actions had no effect as found they had an effect (either positive or negative). And for a minority of the action-outcome pairs, even when most studies found a particular action had positive effects or no effects, at least two studies found the action had negative effects.
As a result of this variation in findings across studies, we cannot conclude with confidence that a mediator action will have a positive (or negative) effect on mediation outcomes, only that the action can have a positive (or negative) effect and, in some instances, could have an effect in the direction opposite from that seen in the majority of studies.
A summary of the research findings for each category of mediator actions and each set of mediation outcomes follows, ending with overall conclusions about which mediator actions, on balance, appear to have a greater potential for positive (or negative) effects on mediation outcomes.
Pressing or Directive Actions
These actions generally either increased or had no effect on settlement, but in some studies were associated with reduced settlement, less joint goal achievement, and more post-mediation adversarial motions being filed. Virtually all studies found mediator pressure on or criticism of disputants either had no effect or had negative effects on disputants’ relationships and perceptions of mediation. Thus, pressing or directive actions have the potential to increase settlement, but they also have the potential for negative effects on settlement and related outcomes and especially on disputants’ perceptions and relationships.
Offering Recommendations, Suggestions, Evaluations, or Opinions
Recommending or proposing a particular settlement, suggesting possible options or solutions, or offering some form of case evaluation or other views about the dispute or its resolution generally either increased or had no effect on settlement. These actions were not related to the personalization of mediated agreements, reaching a consent order, or filing post-mediation adversarial motions or enforcement actions. Offering recommendations, suggestions, evaluations, or opinions had mixed effects on disputants’ relationships and perceptions of mediation, with studies finding positive, negative, and no effects. With regard to attorneys’ perceptions of mediation, these actions generally either had no effect or were associated with more favorable views, with the latter seen especially in early neutral evaluation. Thus, this set of actions has the potential for positive effects on settlement and on attorneys’ perceptions of mediation but has the potential for both negative and positive effects on disputants’ relationships and perceptions of mediation.
Eliciting Disputants’ Suggestions or Solutions
These actions generally increased settlement and also were related to disputants’ greater joint goal achievement, reaching a consent order, and being less likely to file a post-mediation enforcement action, but they were not related to the personalization of mediated agreements or the filing of post-mediation adversarial motions. Eliciting disputants’ suggestions or solutions either had no effect or had positive effects on disputants’ relationships and perceptions of mediation. Thus, eliciting disputants’ suggestions or solutions has the potential to increase settlement and enhance disputants’ perceptions and relationships, with no reported negative effects.
Addressing Disputants’ Emotions, Relationships, or Hostility
Giving more attention to disputants’ emotions, relationships, or sources of conflict generally either increased or had no effect on settlement, and either reduced or had no effect on post-mediation court actions. These mediator actions either had no effect or positive effects on disputants’ relationships and perceptions of mediation. Trying to reduce emotional tensions or control hostility had mixed effects on settlement, with studies finding positive, negative, and no effects; these actions were not examined in relation to disputants’ perceptions. Thus, giving more attention to disputants’ emotions or relationships has the potential to increase settlement and enhance disputants’ relationships and perceptions, but it also has the potential to reduce settlement. Addressing disputants’ hostility has the potential both to increase and to reduce settlement.
Working to Build Trust, Expressing Empathy, Structuring the Agenda, or Other “Process” Actions
Working to build rapport and trust with and between the disputants, expressing empathy, praising the disputants, or structuring the issues and agenda generally either increased or had no effect on settlement. Other process-focused actions and approaches, such as summarizing, reframing, or using a facilitative or nondirective style, had mixed effects on settlement, with studies finding positive, negative, and no effects. These various mediator actions generally either had no effect or had positive effects on disputants’ relationships and perceptions of mediation. Thus, working to build trust, expressing empathy or praise, and structuring the agenda have the potential to increase settlement and enhance disputants’ relationships and perceptions. Other “process” actions have the potential for positive effects on disputants’ perceptions and settlement, but they also have the potential to reduce settlement.
Using Pre-Mediation Caucuses
The effects of pre-mediation caucuses depended on their purpose. When used to establish trust and build a relationship with the parties, pre-mediation caucuses increased settlement and reduced disputants’ post-mediation conflict. But when used to get the parties to accept settlement proposals, pre-mediation caucuses either had no effect or had a negative effect on settlement and post-mediation conflict. Thus, pre-mediation caucuses with a focus on building trust have the potential for positive effects, but those with a focus on substantive proposals have the potential for negative effects.
Using Caucuses During Mediation
Caucuses generally increased settlement in labor-management disputes but had no effect on settlement in other types of disputes, regardless of whether the goal was to establish trust or discuss settlement proposals. Caucusing also was not related to disputants’ joint goal achievement, the personalization of mediated agreements, or whether disputants reached a consent order or filed post-mediation adversarial motions, but disputants who spent more time in caucus were more likely to return to court to file an enforcement action. Caucusing generally either had no effect or had a negative effect on disputants’ perceptions and post-mediation conflict. Thus, caucuses during mediation appear to have the potential to increase settlement in the labor-management context but have the potential for negative effects on disputants’ relationships and perceptions more broadly.
Across the studies reviewed, none of the categories of mediator actions was found to have consistent effects on any of the three sets of mediation outcomes, some actions had different effects on settlement than on disputants’ relationships and perceptions of mediation,5 and a considerable proportion of studies reported no effects. Accordingly, the research does not permit clear conclusions that certain mediator actions will enhance mediation outcomes while others will have detrimental effects.
Looking at the relative potential for positive versus negative effects, while bearing in mind the substantial likelihood of no effects, the following mediator actions appear to have a greater potential for positive effects than negative effects on both settlement and related outcomes and disputants’ relationships and perceptions of mediation:
- Eliciting disputants’ suggestions or solutions
- Giving more attention to disputants’ emotions, relationship, and sources of conflict
- Working to build trust and rapport, expressing empathy or praising the disputants, and structuring the agenda
- Using pre-mediation caucuses focused on establishing trust
Some of the above actions, however, have been examined in only a relatively small number of studies and in only a subset of dispute types, primarily divorce, limited jurisdiction, community, and labor disputes.
The potential effects of other mediator actions appear more mixed. Recommending a particular settlement, suggesting settlement options, and offering evaluations or opinions have the potential for positive effects on settlement and on attorneys’ perceptions of mediation, but these actions have the potential for negative as well as positive effects on disputants’ relationships and perceptions of mediation. Both caucusing during mediation and pressing or directive actions have the potential to increase settlement and related outcomes, especially in labor-management disputes, but pressing actions also have the potential for negative effects on settlement. And both sets of actions have the potential for negative effects on disputants’ perceptions and relationships.
This variation in findings across studies and across sets of outcomes demonstrates that drawing conclusions about the effectiveness of mediator actions based on the findings of a single study or on a single set of outcomes could lead to recommendations not supported by the overall pattern of research findings. The Task Force’s analysis of the studies, in short, suggests that mediators and others should be cautious about broad claims about “the research” showing that any particular mediator action constitutes “best practice.”
Proposed Next Steps and Recommendations
To further the development of a reliable empirical understanding of the effects of mediator actions as well as the creation of links between researchers and the broader mediation community, the Task Force Report lists six proposed steps along with specific recommendations to guide their implementation. The Task Force recommends that two collaborative bodies be established to oversee and implement these steps: a working group under the auspices of the ABA Dispute Resolution Section and a university consortium of mediation researchers. The Section recently created a small group to develop potential mechanisms for following up on these recommendations.
Some of the proposed steps aim at expanding and refining the field’s understanding of the effects of mediator actions using existing research, such as by examining a broader set of studies in mediation and other fields and by undertaking a more nuanced analysis to see how characteristics of the dispute, context, and other factors alter the effects of mediator actions. Other proposed steps involve ways to create future studies whose findings can more meaningfully be compared and aggregated, for example, by developing greater uniformity and consistency across studies in how mediator actions and mediation outcomes are defined and measured, creating more reliable and valid measures, and using more rigorous methodologies.
The Task Force Report also proposes developing and expanding links between researchers and mediation trainers, practitioners, and program administrators to create ongoing collaboration and facilitate the exchange of questions and findings. This includes encouraging greater mediator involvement in research; developing ways to improve dissemination of and access to research findings; and creating mechanisms to incorporate empirical findings into mediation practice, such as through guides for mediator training, performance assessments, quality standards, and feedback mechanisms.
Task Force members see these proposed future steps as essential for the field of mediation to be able to develop a body of empirically derived knowledge that can deepen our understanding of the effects of mediator actions and create ways to incorporate that knowledge into practice. Moving beyond anecdotes to an evidence-based approach will help us improve mediation clients’ experiences and outcomes.
- The full Report is available at https://www.americanbar.org/groups/dispute_resolution.html.
- We limited our inquiry to research findings reported in English.
- Assessing the methodology of each study in detail was beyond the resources of the Task Force. Accordingly, the report of the empirical findings includes all studies reviewed and treats them with equal weight, regardless of how rigorous their research methodology and data analysis were.
- See Tables V.H.1 to V.H. 3 in the Report, supra note 1.
- Too few studies examined the effect of mediator actions on attorneys’ perceptions to compare them to the other outcomes.