Article

The Psychology of Procedural Preference: How Litigants Evaluate Legal Procedures Ex Ante

“Legal authorities can both do their jobs well and create public satisfaction. The key is to have a clear understanding of what people want from the courts . . . . The first issue involved in knowing what people want from the courts is to examine people’s preferences concerning how disputes should be resolved.”

 —Tom R. Tyler, Citizen Discontent with Legal Procedures: A Social Science Perspective on Civil Procedure Reform, 45 Am. J. Comp. L. 871, 876 (1997).

I.     Introduction

The global economic downturn that sent the United States into a severe recession from 2007 to 20091 has wreaked havoc on court systems throughout the country. The recession has forced forty-three states to substantially cut their judicial budgets,2 which is striking given that more than ninety-six percent of all litigation occurs in state courts.3

The waiting time for civil trials in many jurisdictions has dramatically increased—in at least one major metropolitan area, the waiting time for many cases is five years or longer4—and budgets for alternative dispute resolution (“ADR”) programs have shrunk.5 Thus, in the contemporary legal world, many litigants struggle to obtain civil justice.

In this light, lawyers and courts must strive harder to assist litigants in need. But capable assistance requires an understanding of what litigants think about their options. The need for empirical research that elucidates litigants’ perceptions of, and preferences for, procedures—e.g., arbitration, mediation, negotiation, trial—has arguably never been greater. Many courts offering a single alternative to trial choose between mediation and non-binding arbitration.6 But which of these two procedures do litigants prefer?

Court personnel and legislators can use research that clarifies such preferences to design programs that appeal to litigants.7 These programs can have substantial consequences for litigants and court systems alike, by affecting the relative costs of litigation borne by the government, the amount of time necessary to resolve disputes, and the number of cases on court dockets.8 Such research could also help lawyers to better understand litigant psychology in ways that improve how they counsel their clients and predict the preferences of opposing parties.9

Yet despite nearly four decades of empirical research on how laypeople evaluate procedural options, our understanding of litigants’ perceptions of procedures at the inception of their cases—i.e., ex ante—remains somewhat of an enigma. One reason for the gap in our knowledge is the striking uniformity of the methodology used in past research. First, most of the research examining how laypeople perceive and evaluate their procedures—largely falling under the rubric of “procedural justice”10 research—has consisted of laboratory studies of laypeople (typically undergraduate research participants) who evaluate options for resolving hypothetical disputes.11 Such research is important and profound in its own right. But the generalizability of such findings to real-world disputes remains unclear.12 Second, the few field studies on actual civil disputants are also remarkably homogeneous—with the exception of just two published studies, such research has examined the attitudes of real civil litigants only ex post—i.e., after they have experienced a procedure that resolved their dispute.13

By contrast, the research presented here aims to answer an important overarching question: How do civil litigants evaluate procedures shortly after their cases have been filed? To our14 knowledge, this Article reports the first multi-jurisdictional study of how civil litigants whose cases are filed in court assess legal procedures ex ante.15

Part II of this Article discusses the importance of understanding legal procedures from the litigant’s perspective, and describes the motivation behind this large-scale project. Part III synthesizes the existing literature on procedural preferences, and introduces the study’s novel methodology. Part IV presents surprising findings suggesting that, contrary to what some scholars have argued, litigants prefer mediation to most adjudicative procedures ex ante. Other analyses, which examine the relation between the attractiveness of common procedures and demographic, case type, relationship, and attitudinal factors, suggest that only some of the factors previously regarded as important predictors of litigant attraction are significant when evaluated concurrently. Lastly, Part V discusses the implications of these findings for court policy, client counseling protocols, procedural justice, and our understanding of litigant psychology more broadly.

II.     The Importance of Understanding How Litigants Evaluate Procedures Ex Ante: Rationale for the Current Study

Understanding litigants’ ex ante perceptions of procedures is important for philosophical reasons as well as practical ones concerning how courts design ADR programs and how lawyers counsel their clients. First, from a philosophical perspective, parties should be able to “own” their disputes, with their preferences guiding the resolution, because the very concept of “justice” concerns the needs and values of people who bring their conflicts to the legal system.16 Many courts and local bar associations have echoed this sentiment by codifying rules regarding how lawyers should discuss procedures with their clients.17 These rules signal the importance of considering litigants’ subjective perceptions of their options ex ante. From a practical standpoint, insofar as lawyers rely on their own (possibly biased) intuitions about how litigants view procedural alternatives, or on anecdotes about what past clients preferred for their own cases, it is doubtful that they can counsel clients effectively.18 Research on how litigants psychologically process their options is the key to dispelling any misunderstandings that lawyers may have when trying to understand disputes from the litigant’s perspective,19 and explaining options in terms that make sense from that perspective.

Second, by designing programs that reflect disputants’ subjective preferences, court personnel and other policymakers can advance the goals of democratic governance. After all, “[d]emocracy functions as a system in which formal and informal institutions serve the purpose of translating social preferences into public policies [and d]ispute resolution mechanisms are among these institutions.”20 If courts seek to promote values associated with procedural justice, or other values that rely on the subjective satisfaction of disputants, it is necessary to understand disputants’ preferences. Similarly, client counseling protocols that reflect an understanding of litigant psychology regarding dispute resolution can promote more meaningful party participation in the resolution of disputes.

Third, from a practical perspective, courts could enhance public satisfaction with the legal system by designing ADR programs based on research that elucidates litigant preferences. Presumably, litigants in the aggregate would find such programs more attractive, and be more likely to use them (i.e., in voluntary programs) and participate in good faith (in either voluntary or mandatory programs).21 Public sentiment is important because the smooth functioning of the legal system “heavily [depends] on the voluntary cooperation of most citizens.”22 Contemporarily, it is especially critical to consider public sentiment surrounding the legal system because severe reductions in available resources have challenged courts across the country.23

A fourth reason underscoring the importance of litigants’ subjective perceptions concerns the need for court intervention in the regulation of disputes. Empirical research suggests that litigants are less likely to continue a dispute, and more likely to voluntarily comply with the terms of an agreement to resolve it, when they are satisfied with their dispute resolution experience.24 Thus, offering options that litigants find attractive could lead to fewer appeals (when outcomes are produced through trial) and fewer breach-of-contract claims due to noncompliance with a settlement agreement (when the outcomes are derived from a settlement procedure). Either scenario would result in less need for court intervention. Moreover, when people regard the government as offering subjectively attractive and fair procedures, they subsequently demonstrate greater respect for the legal system, and tend to more readily comply with even unrelated laws and regulations.25 Courts surely benefit from such voluntary compliance with the law.

In sum, if courts better understand litigants’ preferences and design their programs accordingly, they could begin to counteract the effect of shrunken budgets on their ability to effectively mete out justice. Even if litigants want their day in court, many cannot afford—financially or emotionally—to wait several years for a trial, and alternatives that subjectively appeal to them could provide some relief. Empirical research could also improve attorneys’ understanding of how litigants perceive procedures, which could enable them to counsel clients more effectively. In this light, the next Part elaborates on what the existing research literature suggests regarding litigants’ procedural preferences, and where gaps in knowledge continue to exist.

A.     What We Already Know About Disputants’ Preferences

Several decades of empirical studies—most of which come from the “procedural justice” research area within psychology—have attempted to explore how laypeople evaluate legal procedures.26 This Part reviews the major findings stemming from laboratory experiments and field research on procedural preferences. It then describes the debate concerning the conclusions that one might draw from these findings. Lastly, the field research that sets the stage for the present research is described in detail.

1.     What Past Research Suggests About Procedural Preferences

Studies on procedural preferences have generally taken one of two forms: the laboratory study and the field study. In the typical laboratory experiment, research participants—typically college students—are randomly assigned a “role” (e.g., the viewpoint of the plaintiff or the defendant) from which to consider the facts of a hypothetical dispute. They then read descriptions of procedures and evaluate the attractiveness of each option for the given dispute.27 Most of this research assesses ex ante perceptions of procedural options.28 In some studies, however, participants engage in simulated procedures and rate them ex post.29

Laboratory research tends to examine only one or two variables of interest within a given study. These variables tend to be experimentally manipulated (e.g., the party’s role in the dispute, the level of evidentiary support for each party’s position, and the nature of the relationship between the parties).30 Despite the richness of the available laboratory research on laypeople’s ex ante preferences, how “generalizable” the findings are to real litigants involved in actual civil litigation remains an open question.31

Field studies complement laboratory research by examining the attitudes of real disputants. In a typical field study, actual litigants evaluate their dispute resolution experiences retrospectively.32 To our knowledge, only two published field studies have examined perceptions of real disputants prospectively (i.e., before a procedure has resolved the dispute).33

The multitude of studies across both the laboratory and field study formats has produced interesting variability in findings. Many laboratory experiments support the idea that people generally prefer more adjudicative procedures (e.g., trial or arbitration) to less adjudicative ones (e.g., mediation).34 Similarly, the famous RAND field study of tort cases found that litigants gave higher procedural fairness and satisfaction ratings to non-binding arbitration and trial than to the less adjudicative settlement conference.35

By contrast, other studies suggest that disputants prefer nonadjudicative procedures. For example, laboratory research by Heuer and Penrod found that participants preferred mediation under some conditions,36 and those confronted with a conflict that had “integrative settlement” potential37 preferred bargaining to both mediation and arbitration.38 Similarly, the often-cited field research on labor grievances by Jeanne Brett, Stephen Goldberg, and their colleagues concluded that disputants favored mediation to arbitration.39 Other field studies, including a study of over 400 cases administered by four major ADR service providers in four different states,40 have reinforced the idea that disputants prefer mediation to adjudicative options.41

2.     Debate over Conclusions from Past Research

What to make of the findings from the hundreds of studies on disputants’ procedural preferences has prompted some debate. Deborah Hensler, for example, has criticized the use of mediation—particularly mandatory mediation—in many courts, asserting that there is no clear empirical evidence that litigants prefer mediation to adjudicative procedures, such as arbitration.42 Others, such as Lisa Blomgren Bingham, have argued that the existing literature suggests that litigants prefer mediation to arbitration.43 Others have argued that the research conclusions seem to vary according to when the studies were conducted, since the historical trajectory of ADR tracks changes in the public’s awareness and opinion of various alternatives to trial.44 Other scholars have suggested that litigant attraction to various procedures may depend on when during the dispute resolution trajectory their perceptions are assessed,45 with some proposing that more adjudicative procedures are preferred ex ante, whereas nonadjudicative procedures are preferred ex post.46

Given the controversy, Hensler has raised an interesting question: “What would a system of dispute resolution based on litigants’ preferences—rather than lawyers’ self-interest or judges’ beliefs about their appropriate role—look like?”47 Despite decades of research on disputants’ preferences, scholars continue to disagree on the precise answer. The lack of clarity notwithstanding, several reliable findings have emerged from the research literature as a whole. For example, it would be difficult to challenge the conclusion that process is critically important to disputants—often just as, if not more, important than outcome.48 It would also be difficult to counter the more specific findings that disputants highly value opportunities for voice, fair treatment by third parties, and disputant control over process, and that these factors heavily influence their evaluation of procedures.49 It is also well established that “[t]he best predictor of subjects’ preference ratings . . . is their rating of the fairness of the various procedures, with greater preference expressed for those procedures deemed most fair.”50

Nevertheless, when it comes to whether litigants—at the start of their cases—prefer either nonadjudicative or adjudicative procedures, clarity remains elusive. Since most studies examining the ex ante perspective were conducted in laboratory settings and involved the examination of preferences for hypothetical disputes, their findings may not generalize to the real world.51 It is also not clear whether civil litigants have a preference between mediation and non-binding arbitration, the two ADR procedures that courts most commonly offer.52 To the extent that litigants have a preference between these two procedures, courts choosing between them could benefit from knowing which is favored, or under which conditions litigants prefer either option.

3.     Review of the Two Field Studies on Litigants’ Ex ante Preferences

Thus far, only two published studies have explored how actual civil disputants assess legal procedures ex ante. The first, by Lamont E. Stallworth and Linda K. Stroh, investigated Equal Employment Opportunity Act disputes pending before the Illinois Human Rights Commission, and compared fact-finding, mediation, and binding arbitration.53 Their findings suggest that litigants are more interested in mediation than binding arbitration.54 Although the study examined ex ante perceptions, many of the participants completed surveys only after an initial investigation into their case was made, and this process often took several years.55 Thus, they “could not ascertain disputants’ willingness to use ADR at the very early or initial filing stage.”56

The second ex ante field study, which served as the pilot study for the present research, differed from Stallworth and Stroh’s study in important ways. First, it was a longitudinal project designed to compare ex ante and ex post evaluations.57 The ex ante phase involved mailing surveys to litigants within two weeks of when the cases were filed.58 Thus, compared to the Stallworth and Stroh study, litigants’ perceptions were assessed much earlier in the dispute resolution trajectory. Second, although this project was like the Stallworth and Stroh study in that it surveyed disputants from only one jurisdiction,59 it included litigants with a broader range of case types.60 Third, rather than examine attitudes toward concrete legal procedures such as mediation or trial, the ex ante survey tested the hypothesis that litigants involved in active litigation tend to cognitively sort through procedural options based on how much control they offer to the litigants themselves versus third parties. To test this hypothesis, participants evaluated options for the core characteristics of procedures (i.e., outcome, process, and substantive rules)—called “feature options”—and rated the attractiveness of each for their particular dispute. For example, they assessed sets of options pertaining to the outcome (e.g., who would make the final decision and whether that outcome would be advisory or binding), how the process would evolve (e.g., how informal the process would be; whether disputants could express themselves conversationally or only in response to questions posed by others), and the substantive norms or rules that would be used to resolve the dispute (e.g., whether the law would automatically apply or the parties could opt to use other standards). Although variations of this feature-based approach had been common in laboratory studies, this was the first published study that applied this paradigm to real litigants.61

The findings from this pilot research suggest that litigants tend to assess procedural options ex ante by categorizing them in terms of how much control they offer to third parties as opposed to the litigants themselves. This pattern corresponds to what past researchers have found in laboratory research.62 In addition, the study found “that older disputants were less attracted to third party control than their younger counterparts,”63 and those “involved in contract disputes tended to prefer disputant control more than those involved in other kinds of conflicts.”64 The results also revealed that litigants who opposed a collective (i.e., a company, organization, or similar body) were “less attracted to feature options offering disputant control than those” who opposed an individual.65

In addition, litigants’ ex ante attraction for adjudicative versus nonadjudicative procedures did not predict the type of procedure (i.e., adjudicative versus nonadjudicative) that was used to ultimately resolve the dispute. It is important to note that the court from which litigants were recruited did not offer both mediation and arbitration, in addition to trial, for the very same types of cases. The fact that many, if not most, cases were eligible for a single court-connected ADR procedure may have skewed litigants’ perceptions of their options, as well as their choice of which procedure to use.

Given that only two relevant ex ante field studies exist, laboratory experiments and field research focused on understanding disputants’ ex post evaluations constitute almost the entire body of research pertaining to how disputants assess legal procedures. Because the existing field research focuses on ex post evaluations, how actual civil litigants assess procedures ex ante remains a bit of an enigma. How do litigants evaluate specific procedures shortly after their cases have been filed? Of particular interest is how they compare mediation to non-binding arbitration. Another question concerns litigants’ expectations—that is, what procedures do they expect to use? Understanding how litigants’ attraction for procedures relates to their expectations for using those same procedures can shed light on whether they believe their own predilections will drive how the resolution of their dispute will unfold, or whether they believe other factors will play a role.

Yet another set of open questions concerns whether litigant attraction for the various procedures depends on the type of case that is involved, or on demographic, relationship, or attitudinal factors. Laboratory research on ex ante perceptions suggests that many factors can influence the way in which people view their dispute resolution options. These previously examined factors include the strength of the case,66 the relationship between the parties,67 the party’s role in the case (i.e., defendant or plaintiff),68 the type of dispute,69 the litigant’s culture,70 race, or ethnicity and gender.71

Armed with answers to these key questions, court administrators could channel their limited resources towards the kinds of procedures that appeal to disputants. In addition, lawyers could better understand litigant psychology regarding their options, allowing them to better counsel their clients and better predict the preferences of opposing parties.

III.     Method

This study surveys litigants from three distinct state court systems. To our knowledge, it is the first multi-jurisdictional field study of litigants’ ex ante procedural preferences. It examines preferences within a “laboratory-like” environment that holds constant as much as possible the court-connected options that litigants could consider. Specifically, the courts were selected precisely because they offered both mediation and non-binding arbitration, in addition to trial, for the same types of cases. Compared to earlier studies, it examines evaluations of a wider variety of procedures, and explores a larger set of factors that might predict attraction to procedures. Thus, the design of the project affords the possibility of superior generalizability of ex ante perceptions compared to other published research.

A.     Study Courts

Participants were recruited from general jurisdiction state trial courts (the “study courts”) located in three states:

Third Judicial District Court, Salt Lake City, Utah (“Utah Court”). The Utah Court is located in Salt Lake City. With a 2010 population of 1,124,197, it has the most populous judicial district in the state.72 In 2010, litigants filed 55,074 civil cases in the Utah Court.73

Superior Court of Solano County, California (“California Court”). The California Court is located in northern California, with courthouses in Fairfield and Vallejo. Solano County has an estimated population of 413,334.74 For the 2010–2011 fiscal year, the Solano County Court received 13,910 civil filings.75

Fourth Judicial District, Multnomah County, Oregon (“Oregon Court”). The Oregon Court is located in Multnomah County, Oregon’s geographically smallest but most populous county.76 Multnomah County had 735,334 residents in 2010,77 which places its population between Utah’s Third District and California’s Solano County. In 2010, the Oregon Court received 18,203 civil case filings.78

See Appendix B for additional details concerning each court and its programs.

B.     Materials

The survey collected basic demographic information (e.g., gender, race, age group) about the litigants as well as some details about their cases (e.g., whether they were the plaintiff, defendant, or both,79 whether they were involved in the case as an individual, or were representing a company, organization, or group, whether they had been involved in a previous case as a plaintiff or defendant, and which parties (if any) were affiliated with an insurance company that had an interest in the outcome of the case). Case type (e.g., contract, property, personal injury) information was also collected. Litigants indicated whether they knew or had a relationship with the opposing party prior to the filing of their case, and rated the importance of having a relationship with the opposing party in the future (1 = not at all important; 5 = extremely important). Other attitudes were also assessed: participants provided a 0–100% chance estimate of how strong they believed their case to be (“If you go to trial for this case, what do you think your chances are of ‘winning’?”), and rated their “impression of the court where this case has been filed” (1 = extremely negative; 9 = extremely positive). For further information about the participants and their case types, see infra Tables A and B respectively.

Another set of questions assessed how attractive litigants found the following legal procedures: (1) Attorneys Negotiate without Clients, (2) Attorneys Negotiate with Clients Present, (3) Mediation, (4) Judge Decides without Trial, (5) Jury Trial, (6) Judge Trial, (7) Binding Arbitration, and (8) Non-binding Arbitration. Litigants read a brief description of each procedure (see Appendix D) to ensure construct validity and then rated how attractive they perceived each to be for their case (1 = not attractive at all; 9 = extremely attractive). They rated how likely they thought they were to use each of these procedures “at some point for this case” (1 = not at all likely; 9 = extremely likely). In free-response form, they also indicated how they would decide which procedure(s) to try (“What factors will you consider or weigh when deciding which procedure(s) to use?”). The survey also included questions for a separate project.

Three versions of this survey were created, each using different orders of the same questions.80 The RANDBETWEEN function in Microsoft Excel was used to determine which version each litigant would receive.

C.     Participant Recruitment

For six two-week periods, between May 2010 and May 2011, the research team identified litigants who met the following study criteria in each study court: the litigant’s case must have been filed during the two-week period, and the case type must have been eligible for the court’s mediation and non-binding arbitration programs, as well as trial. The types of cases that met the latter eligibility requirement are reported in Appendix A.81 The list of litigants who met the study criteria was called a “case pull.”82 Each court had its own case pull for each time period.

A predetermined number of litigants were recruited from each case pull. For the Oregon and Utah Courts, the RANDBETWEEN function in Excel was used to randomly select which litigants to invite. Because the California Court had significantly fewer case filings relative to the other two courts, the first-listed defendant and plaintiff for each case were eligible for recruitment.

After completing this modified random-selection process, the team researched addresses for the selected litigants. When the court did not provide litigant contact information, this step typically involved significant research83 because we did not want to risk data contamination by sending the surveys to the attorneys to distribute to their clients. Appendix C describes the address research methodology.

Research assistants rated their “level of certainty” for each address they found (1 = little to no certainty that the contact information for the correct litigant was found; 5 = absolutely certain that contact information for the correct litigant was found).84 To ensure wise use of financial resources, the research team then excluded addresses with certainty ratings below 3.85 To avoid dependency issues in the data, litigants were omitted as needed to avoid sending surveys to more than one plaintiff and defendant in each case, and in instances where the same litigant had multiple cases filed during the study period, to avoid having more than one survey completed by the same litigant within a given case pull.86 Ultimately, the combination of the mailing lists resulting from the case pulls contained more defendants than plaintiffs, suggesting that there were relatively more defendants than plaintiffs listed as litigants during the recruitment period.

The research team mailed surveys and pre-paid return envelopes to litigants within three weeks of the date their cases were filed. An introductory letter and consent form explained that those returning the survey would receive compensation.87

D.     Participants

The dataset (“Litigant Ex Ante Perception Dataset”) contains data from 413 litigants, reflecting a 10% response rate.88 The sample includes litigants with mailing addresses from 19 states; 7.02% of litigants had addresses from outside of the states where the study courts were located.

Table A: Participant Information

Shest 1

Table A: Participant Information (continued)

Shest 2

Table A: Participant Information (continued)

Shest 3

Note: N = 413. Missing data indicates litigants for whom a response to the question was not obtained. Party Type and Opposing Party Type calculations include participants (n = 4 and n = 7, respectively) who indicated that more than one type applied to their case.

*Those indicating that they had been a plaintiff or defendant before were asked how many times they had acted in such a capacity in previous litigation. The modal response to both questions was 1 [(Mplaintiff = 100.35, SD = 419.16); (Mdefendant = 8.63, SD = 26.88)].

**Calculations are based on a question concerning whether the litigants chose their lawyer for this case. Litigants who indicated “don’t have a lawyer” were counted as not having a lawyer; those who replied “yes” or “no” were counted as having a lawyer; litigants who indicated in any open-ended question that they were a lawyer or had a law degree had their responses modified accordingly.

Table B: Case Types

Shest 4

Note: N = 413. When litigants reported multiple case types, they were given the designation of having “two or more case types” rather than being included in individual case type categories. For example: if a litigant reported his case as involving both personal injury and medical malpractice, it was included under the “two or more case types” category only.

Table C: Variables Used as Predictors

Shest 5

This exploratory study was designed to investigate key issues regarding the ex ante perspective. First, a statistical analysis ascertained how litigants evaluated the attractiveness of various legal procedures. The comparative nature of this analysis provided information about relative preferences. Further analyses examined the procedures to which litigants gave extreme ratings, thereby illuminating the procedures that tended to polarize attitudes. Next, an analysis was performed to assess litigants’ expectations regarding the use of each procedure at some point in the resolution of their disputes. Subsequent statistical tests determined whether there was a significant difference between litigants’ attraction for each procedure and how much they expected to use it. Finally, regression analyses were used to determine whether demographic, case type, relationship, or attitudinal variables affect litigants’ attractiveness ratings for each procedure.

IV.     Results

A.     Relative Attractiveness of Procedures

Litigants rated how attractive they perceived each procedure to be for their particular dispute (1 = not attractive at all; 9 = extremely attractive). A Hierarchical Linear Model (“HLM”) was used to assess differences in the attractiveness (i.e., relative preferences) ratings for the various procedures.89 In this instance, conducting the HLM analysis involved choosing one procedure as the benchmark (“reference group”) and then comparing ratings for that procedure with the ratings for each of the other procedures. The Judge Trial was chosen as the reference group because it is the default procedure for any filed case, unless a litigant initiates some alternative.90

As catalogued in Table D, the mean (i.e., “average”) attractiveness of the Judge Trial was 5.72 (the “intercept”). Two procedures received ratings that were statistically equivocal to that of the Judge Trial: Attorneys Negotiate with Clients Present and Mediation. The Judge Trial was rated significantly higher in attractiveness than all of the other examined procedures. Descriptively, the procedure with the highest mean attractiveness was Attorneys Negotiate with Clients Present, and the procedure with the lowest mean attractiveness rating was Binding Arbitration. Figure 1 illustrates the mean attractiveness of the different procedures.

Table D. Estimates of the Prediction of Attractiveness of Procedures: Judge Trial Reference Group

Shest 6

*** p < .001. ** p < .01. * p < .05. Note: The intercept represents the mean attractiveness rating for the reference group (i.e., Judge Trial). For all regression tables included in this Article, B indicates the estimated unstandardized coefficient, SE indicates the standard error of that coefficient, t indicates the test statistic.

Figure 1: Procedure Attractiveness

 Shestowsky 1

Figure 1. Mean litigant attraction ratings of legal procedures. Error bars are SEs of the ratings. Reported above each bar is the average litigant rating for that procedure. Litigants rated their attraction to each procedure using a 9-point Likert scale.

One motivation of this project was to compare litigants’ attraction to common forms of court-connected ADR—specifically, mediation and non-binding arbitration. Given the debate regarding whether or not litigants prefer mediation to more adjudicative procedures, analyses comparing the attractiveness of Mediation with that of Binding Arbitration, the Jury Trial, and the Judge Trial were also of interest. Thus, an additional HLM analysis was conducted using Mediation as the reference group. As catalogued in Table E, the mean attractiveness of Mediation was 5.94 (the “intercept”). The analysis revealed that litigants significantly preferred Mediation to Non-binding Arbitration. It also revealed a complicated picture regarding how Mediation compares to adjudication: litigants significantly preferred Mediation to all forms of adjudication except for the Judge Trial. This analysis also demonstrated that litigants significantly preferred Mediation to the Attorneys Negotiate without Clients option, but did not have a preference between Mediation and the Attorneys Negotiate with Clients Present.

An additional analysis comparing attraction for the two forms of arbitration revealed that litigants liked Binding Arbitration significantly less than Non-binding Arbitration.91 A final analysis compared the two forms of negotiation; litigants preferred to negotiate with the parties alongside their attorneys rather than having the attorneys negotiate without the parties.92

Table E: Estimates of the Prediction of Attractiveness of Procedures: Mediation Reference Group

Shest 7

*** p < .001. ** p < .01. * p < .05. Note: The intercept represents the mean attractiveness rating for the reference group (i.e., Mediation).

B.     Comparisons of Extreme Attitudes Regarding Procedures

Given that litigants rated attractiveness using 9-point Likert scales, the first and ninth categories (1 = not attractive at all; 9 = extremely attractive) are of special interest because they represent extreme attitudes regarding the procedures. Chi-square tests93 were performed to assess whether litigants were significantly more likely to rate any of the procedures as “not attractive at all” versus “extremely attractive.” This information is not captured by analyses that focus on the mean ratings. Table E catalogues the results.

There were no significant differences between the observed frequencies for the extreme-attractiveness ratings of Attorneys Negotiate without Clients, Judge Decides without Trial, the Judge Trial, and the Jury Trial, suggesting that litigants were as likely to rate these procedures as “extremely attractive” as they were to rate them “not at all attractive.” By contrast, significantly more litigants rated Attorneys Negotiate with Clients Present as “extremely attractive” as opposed to “not at all attractive,” and this was also true of Mediation. The opposite was true for both forms of arbitration—significantly more litigants rated both Binding Arbitration and Non-binding Arbitration as “not at all attractive” compared to “extremely attractive.”

Table F. Frequencies of Extreme Attractiveness Ratings for ProceduresShest 8

*** p < .001. ** p < .01. * p < .05. Note: df = 1 for all analyses.

C.     Expected Use of Procedures

Litigants also rated each procedure in terms of “how likely [they were] to use them at some point for this case” (1 = not at all likely; 9 = extremely likely). HLM analysis was used to assess differences in the means of the expected likelihoods of using different procedures, using the Judge Trial as the reference group.

As shown in Table G, the mean estimated use of the Judge Trial was 5.05 (the “intercept”). This perceived probability was significantly higher than the perceived probability of using the Jury Trial, Binding Arbitration, and Non-binding Arbitration. The litigants did not view the likelihood of using the Judge Trial as significantly different from the likelihood of using any of the other procedures. Descriptively, the procedure with the lowest expected use was Binding Arbitration and the procedure with the highest mean expected use was Attorneys Negotiate without Clients. Figure 2 illustrates the mean expected use of each procedure.

Table G. Estimates of the Prediction of Use Based on Different Procedures

Shest 9

*** p < .001. ** p < .01. * p < .05. Note: The intercept represents the mean attractiveness rating for the reference group (i.e., Judge Trial).

Figure 2: Expected Use of Procedures

 Shestowksy 2

Figure 2. Mean litigant expected-use ratings for legal procedures. Error bars are SEs of the ratings. Reported above each bar is the average litigant rating for that procedure. Litigants rated their expected use of each procedure using a 9-point Likert scale.

D.     Attractiveness of Procedures Compared to Expected Use

To determine whether litigants’ attraction to each procedure differed from their expectation regarding whether they would use that procedure at some point in the resolution of their dispute, paired t-tests were conducted to test for differences between the attractiveness and expected use ratings for each procedure. Each comparison was statistically significant, except for the Attorneys Negotiate without Clients option.94 Moreover, each difference was in the same direction, suggesting that litigants tended to like each procedure more than they thought they would use it. Figure 3 catalogues this comparison.

Figure 3: Procedure Attractiveness vs. Expected Use

 Shestowsky 3

Figure 3. Comparison of mean litigant attraction to, and expected use of, legal procedures. Error bars are SEs of the ratings. Litigants rated their attraction to, and expected use, of each procedure using 9-point Likert scales.

E.     Predictors of Attractiveness of Specific Procedures

A major question of interest was how demographic, case type, relationship, and attitudinal variables relate to litigants’ attraction to each procedure. To explore this issue, simultaneous multiple regression analyses were conducted.95 The analyses included each procedure as the outcome variable, and a series of demographic, case type, relationship, and attitudinal variables as predictors.96

Attorneys Negotiate without Clients97

Companies liked the option of Attorneys Negotiate without Clients significantly more than individual litigants did. In addition, the more positively litigants viewed the court, the more attracted they were to this procedure, and vice versa. By contrast, there was a significant, negative relation between litigants’ perceptions of winning their case at trial and their attraction to this procedure. Specifically, the more confidence they had in a trial win, the less they liked this procedure, and vice versa. Finally, litigants who had a previous relationship with the opposing party were significantly less attracted to this procedure compared to those who did not. Appendix F reports the results.

Attorneys Negotiate with Clients Present98

Compared to those whose cases involved personal injury matters only, those with a case type of “other” rated the option of Attorneys Negotiate with Clients Present as significantly more attractive. Also, litigants who indicated that an insurance company had an interest in the case were significantly more attracted to this procedure compared to those whose cases did not involve an insurance company. Moreover, how much litigants valued a future relationship with the other party was associated with how much they liked this procedure. Specifically, the more they desired a relationship with the other party, the more attracted they were to this procedure, and vice versa. By contrast, there was a significant, negative relation between age group and attraction: litigants in older age groups liked this procedure less than those in younger age groups. Compared to individual litigants, groups and organizations were significantly less attracted to this procedure.  Appendix G reports the results.

The Jury Trial99

Litigants found the Jury Trial more attractive when they faced an opposing party that was a group or organization (compared to an individual). There was also a significant, positive relation between litigants’ estimate of a trial win and how much they liked the Jury Trial: the more highly they estimated a trial win, the more they liked the idea of a Jury Trial for their case, and vice versa. However, litigants found the Jury Trial less appealing when they were involved in a case that involved only property issues (compared to only personal injury issues), were female, or filed their case in California (as opposed to Oregon). Appendix H reports the results.

The Judge Trial100

Litigants whose cases involved two or more case types rated this procedure as significantly more attractive compared to those whose cases concerned only personal injury matters. Attraction to the Judge Trial was also significantly positively related to litigants’ perceptions of their likelihood of winning at trial and their perceptions of the court where their case was filed. Specifically, the more confident they were of a win at trial, the more they liked the Judge Trial, and vice versa. Similarly, the more favorably litigants viewed the court where their case was filed, the more they liked the idea of having a Judge Trial, and vice versa. Appendix I reports the results.

Judge Decides Without Trial101

The factors that predicted attraction for this option mirrored the pattern observed for the Judge Trial. Litigants whose cases involved two or more case types rated this procedure as significantly more attractive compared to those whose cases concerned only personal injury matters. Litigants’ attraction to this procedure had a significant, positive relation with their perceptions of winning their case, and their attitudes toward the court where their case was filed. Specifically, the more litigants expected a trial win, or the more favorably they regarded the court where their case was filed, the more they liked the idea of having a Judge Decide without Trial. Appendix J reports the results.

Binding Arbitration102

Binding Arbitration was more appealing to litigants who were acting as both a plaintiff and defendant in their case (compared to those acting as a plaintiff only), were repeat-litigants103 (compared to first-time litigants), or were opposing a company (versus an individual). Gender was also a significant predictor of attraction to this procedure: women liked Binding Arbitration less than men did. Appendix K reports the results.

Mediation and Non-binding Arbitration

The regression model did not account for a significant proportion of the variance in attractiveness ratings for Mediation104 or Non-binding Arbitration.105 While one cannot infer on this basis that all of the included predictors are unrelated to attractiveness ratings for these procedures, it does suggest that many of them are. More importantly, it means that the present model, given its specification, is unable to further evaluate these possible relations.

 

table h

V.     Discussion

The analyses examined the ex ante dispute resolution perceptions of civil litigants. It is important to note from the outset that the conclusions drawn herein do not necessarily generalize to ex post evaluations because the decision-making processes that people engage in with respect to pre- and post-experience assessments may differ.106

A.     Preferences for Procedures

Analyses revealed that litigants liked the Judge Trial significantly more than all examined procedures except for Mediation and Attorneys Negotiate with Clients Present (both of which had attractiveness ratings that did not statistically differ from that of the Judge Trial).107 Similarly, litigants liked Mediation more than all other procedures except for the Judge Trial and Attorneys Negotiate with Clients Present (both of which had attractiveness ratings that did not statistically differ from that of Mediation). From this pattern, in conjunction with the fact that Attorneys Negotiate with Clients Present had the highest mean attractiveness rating, we can conclude that litigants preferred the Judge Trial, Mediation, and Attorneys Negotiate with Clients Present to all other examined procedures.108 Within this group of favored procedures, they did not have a  statistically significant preference.

One noteworthy aspect of this pattern is that litigants preferred the Judge Trial to the Jury Trial. The reason for this preference is unclear. One possibility may relate to perceptions about adjudicative procedures popularized in mainstream American media. For instance, litigants might prefer to have a judge as the fact-finder in their cases as a result of unfavorable depictions of jury trials, both in media coverage of sensational trials and advertising campaigns by organizations urging “tort reform.”109 Alternatively, litigants could have believed that the waiting time for the Judge Trial would be less than for a trial by jury, or that a jury trial would cost them more financially or emotionally.110 Future research should try to shed light on the reasons underlying this greater enthusiasm for the Judge Trial.

This pattern also suggests that litigants preferred Mediation and Attorneys Negotiate with Clients Present to all adjudicative procedures, except for the Judge Trial. This constellation of preferences lends support to previous scholarship concluding that litigants tend to prefer nonadjudicative procedures.111 However, the presence of the Judge Trial among the more preferred options impedes general conclusions.

The data also revealed that litigants were more interested in negotiations that would include the attorneys as well as the clients to negotiations that would involve the attorneys only. They also liked Mediation112 as much as the former kind of negotiation, but significantly more than the latter type of negotiation. This finding—along with the fact that litigants preferred Mediation to most adjudicative procedures—suggests that litigants want to be present for, and have the option to informally participate in, the resolution process. This finding may come as a surprise to attorneys who assume that they should conduct settlement discussions on their own.113 Although there might sometimes be strategic reasons for excluding litigants from settlement discussions, lawyers should anticipate a desire on the part of clients to observe or participate in the discussions themselves, and counsel clients on the advantages and disadvantages of that option in light of their particular case.

The study also found that litigants clearly preferred Mediation to Non-binding Arbitration.114 This result has important implications for court ADR programs. Some research suggests that voluntary programs tend to have lower usage rates compared to mandatory ones.115 Court administrators who aim to increase litigants’ use of their voluntary programs will want their ADR option to be more appealing than trial itself, ex ante. The present study suggests that litigants would find Mediation—but not Non-binding Arbitration—significantly more appealing than both the Judge Trial and the Jury Trial.116 Thus, for courts with voluntary ADR programs, especially ones that offer only one ADR procedure, mediation might be particularly useful for attracting litigants. Insofar as good faith participation in a settlement procedure might be associated with its ex ante appeal, courts that sponsor mandatory ADR might also benefit from offering Mediation. Future research might investigate litigants’ perceptions of other court-connected procedures, such as Early Neutral Evaluation and judicial settlement conferences.117

Other interesting insights into litigant psychology emerged when the highest and lowest attractiveness ratings for the different procedures were compared.118 In some cases, the results reinforced interpretations made on the basis of the mean attractiveness ratings. Specifically, with regard to Attorneys Negotiate with Clients Present and Mediation—the mean ratings for which placed them among the most preferred options—if litigants held an extreme view of either procedure, it was significantly more likely that they were extremely enthusiastic in favor of them. Moreover, the opposite was true for Binding and Non-binding Arbitration, which, descriptively, had the lowest mean attractiveness ratings. If litigants had an extreme opinion about either of these procedures, it was more likely that they were extremely opposed to using them for their case. Lawyers should be mindful of these findings when counseling clients because they suggest that some litigants may have polarized views towards procedures that have an “average” rating when evaluated in the aggregate (e.g., Non-binding Arbitration’s mean attractiveness rating was 4.65 out of 9).

For the other procedures, the number of litigants giving the highest possible rating did not differ from the number giving the lowest possible rating. In colloquial terms, the litigants who held extreme attitudes about the attractiveness of these procedures were just as likely to “love” them as to “hate” them.

Notably, the Judge Trial was absent from the “significantly more litigants love this procedure than greatly dislike it” category. And yet this procedure scored relatively high marks on attractiveness when mean ratings were compared. It is possible that litigants did not have a strong emotional reaction to the Judge Trial because they recognized that it was the default option for their case. The expected use data, which revealed that litigants viewed the Judge Trial as more likely for their case than many other procedures,119 support this interpretation.

It is interesting to hypothesize why litigants felt strongly attracted to certain procedures. Commonalities not traditionally considered among the Attorneys Negotiate with Clients Present, Judge Trial, and Mediation options (which were found to be statistically equivocal in attractiveness)120 might bind these procedures together in the minds of litigants. Perhaps litigants prefer procedures in which they believe they will be present and able to participate in a civil, dignified manner—and, if a decision-maker is involved, that it will be a Judge.121 Alternatively, a sense of democracy might cluster these procedures together. Litigants might be favorably disposed towards the Judge Trial because judges are either democratically elected or representatively appointed. Similarly, mediations and negotiations that include the parties might invoke the self-determination aspect of democracy. The former interpretation seems more likely in light of the fact that litigants were significantly more attracted to the Judge Trial than the Judge Decides without Trial option (which was described in the survey as including a judge but rarely including the litigants).122

B.     Expected Use of Procedures

This study determined litigants’ expectations regarding the possibility of using each procedure for their dispute. Descriptively, litigants felt they were least likely to use Binding Arbitration.123 This finding should be interpreted in light of the fact that the sample was composed of cases filed in court, and therefore presumably not subject to pre-dispute arbitration agreements which normally direct disputes to this type of arbitration. Thus, the data reflect litigants’ expectations regarding post-dispute election of Binding Arbitration, and those expectations were not high.

Litigants also thought they were significantly more likely to use the Judge Trial than the Jury Trial, Binding Arbitration, or Non-binding Arbitration.124 In addition, they did not perceive the likelihood of using the Judge Trial for their case as statistically different from the likelihood of using Attorneys Negotiate without Clients, Attorneys Negotiate with Clients Present, Mediation, or Judge Decides without Trial.

It is interesting to compare these expectations with general usage statistics from state courts. Litigants believed that the Judge Trial was more likely for their case than the Jury Trial, which meshes with statistics indicating that bench trials are more common than jury trials for civil cases filed in state courts.125 By contrast, litigants believed that the Judge Trial was as likely, or more likely, than every settlement procedure that was studied. This perception is not what one would expect in light of statistics suggesting that cases rarely proceed to trial.126 In light of this finding, why litigants believe that the Judge Trial is so likely deserves further investigation. One possible psychological interpretation is that, because litigants indicated their perceptions within a few short weeks after their cases were filed, the newness of the lawsuit might have led them to believe that even if they were to attempt settlement, they would face the default Judge Trial in the long-run.127

It would be interesting to follow these cases over time to discern whether litigants realized their expectations.128 Either way, this trend in their predictions suggests that lawyers should consider engaging their clients in early discussions about how uncommon trials are in the contemporary legal landscape.

C.     Attractiveness of Procedures Compared to Expected Use

The findings regarding litigants’ expected use of procedures become even more intriguing when attractiveness ratings are compared to expected use ratings for each procedure.129 For every option—except Attorneys Negotiate without Clients—litigants liked procedures significantly more than they expected to use them. This pattern suggests that litigants recognize that factors other than their own attraction to a procedure will play a role in determining which procedures they will experience. This finding resonates nicely with those of the pilot study, which revealed that litigants’ ex ante attraction for adjudicative versus nonadjudicative procedures did not predict the procedural type that was ultimately used to resolve their disputes.130

Only one procedure deviated from the trend: Attorneys Negotiate without Clients. For this option, mean attraction did not differ from mean expected use.131 This particular deviation makes sense because Attorneys Negotiate without Clients is the only examined procedure that does not require the consent or participation of anyone other than the litigant’s own agent (i.e., the lawyer). This distinguishing feature presumably allows litigants to maintain a level of control over the amount of engagement with the opposing party’s lawyer, which in turn may explain why expectations of use for this procedure were on par with how much litigants liked it.

If factors other than litigants’ attraction to procedures affect their perceptions of how likely they are to use those procedures, a logical follow-up inquiry is what those other factors might be. An open-ended question asked litigants to describe the factors they would consider or weigh when deciding which procedures to use. The modal response—mentioned by 20.3% of the litigants—concerned input from their lawyers (e.g., “I would rely on counsel”; “I will ask my attorney which to choose”).132 This type of response highlights the possibility that the link between preferences for, and actual use of, procedures might be understood in light of an “interaction” between litigant and lawyer preferences. If that is the case, to encourage democratic lawyering, lawyers should manage this early deference to the lawyer’s perspective by making a concerted effort to understand the client’s goals and interests before sharing their own views on the various procedures.

D.     Predictors of Attractiveness of Specific Procedures

A major goal of this project was to determine the demographic, case type, relationship, and attitudinal factors that predict litigant attraction to the various procedures. Importantly, when all of the factors were evaluated at once, the resulting model did not significantly predict attraction to either Mediation or Non-binding Arbitration, the two most commonly offered court-connected ADR procedures. This pattern bodes well for courts wanting to ensure that the decision to offer one of these will not inadvertently favor the predilections of a subset of the litigant population. Overall, only a few factors predicted attraction for the other procedures. A few significant findings from this broad set of analyses are especially worthy of discussion.

1.     Repeat Litigants Liked Binding Arbitration More than First-Time Litigants

Binding Arbitration was the only procedure for which attraction was significantly related to past litigation experience.133 Repeat litigants liked Binding Arbitration more than their first-time counterparts.134 Perhaps repeat litigants sense that Binding Arbitration awards tend to favor repeat players.135 It is also possible that repeat litigants want to avoid the painfulness of protracted discovery—which is generally less intense in arbitration than trial136—and the threat of an appeal following a trial. To the extent that repeat players might like Binding Arbitration more than first-time litigants because they are better acquainted with the aforementioned benefits of this procedure relative to trial,137 lawyers might attempt to even the information playing field by having early discussions about the possible advantages of Binding Arbitration with clients who are new to litigation, even when their cases are filed in court.

The comparative benefits of Binding Arbitration may be mitigated in large commercial disputes, which could explain why companies did not like Binding Arbitration as much as individual litigants did. Such disputes tend to introduce costs traditionally associated with “big case” litigation.138 What is curious is that litigants who opposed a company liked Binding Arbitration more than litigants who opposed an individual. This result is surprising given the wealth of bad press concerning consumer and employment arbitration, which typically involves cases wherein an individual opposes a corporation.139 One possible reason for this finding might be that litigants who opposed a company regarded this procedure as more expected—for whatever reason, be it normative or through the application of some rule—than those who opposed an individual. To explore this interpretation, the expected use ratings for Binding Arbitration of those opposing an individual were compared to ratings from those opposing a company. Indeed, when the opposing party was a company, litigants thought that Binding Arbitration was significantly more likely.140

2.     Confidence in Trial Win Was Associated with Attraction to Court-Related Adjudicative Procedures

The results also revealed that litigants’ subjective estimates of a trial win were related to how attracted they were to the trial-related options that were most adjudicative: Judge Decides without Trial, the Jury Trial, and the Judge Trial.141 Specifically, the more confident litigants were of a trial win, the more attracted they were to these procedures, and vice versa.142 One interpretation of this pattern is that the more certain litigants felt about the strength of their case, the more they trusted jurors and judges to view their case the same way, and vice versa.

The only other procedure significantly associated with confidence ratings was Attorneys Negotiate without Clients. The more confident the litigants were of a trial win, the less they liked this option, and vice versa. The more they believed they would win outright in court, the less interested they were in a negotiation that might open the door for compromise, if they would not be present in that negotiation to interact with the other party or directly influence the outcome.143 The fact that litigants’ predicted success at trial was not associated with how favorably they regarded the other procedures—including “trial-like” Binding Arbitration—suggests that they were more agnostic about whether these options would produce results that reflected how strong they believed their case to be.

From a psychological perspective, this attraction to court-related adjudicative procedures as a function of litigants’ confidence in their case might reflect an egocentric bias.144The egocentric bias, much studied in the litigation area, is observed when individuals construe information in a self-serving way. This bias can lead litigants to believe their case is much stronger than it is. In the present study, 57%145of the litigants thought they had at least a 90% chance of winning at trial, and 24% believed they had a 100% chance. Only 16% thought they had at most a 50% chance of prevailing. The fact that higher confidence was associated with a greater desire for time-consuming and expensive procedures such as jury and judge trials highlights the importance of lawyers having early discussions about what motivates their clients’ desire for trial. Early conversations about the possible weaknesses in their clients’ case, the financial and emotional costs associated with trial, and the waiting time for trial in their jurisdiction might offer litigants a broader and more realistic perspective from which to consider their initial procedural preferences.

3.     Women Liked the Jury Trial and Binding Arbitration Less than Men

Another intriguing finding that emerged was that women liked the Jury Trial and Binding Arbitration less than men did.146 In fact, this was the only instance in which gender was found to be a significant predictor of attraction to procedures. In light of research suggesting that women favor conflict avoidance and minimization,147 this female distaste for more adversarial procedures like Binding Arbitration and Jury Trials is understandable. What is surprising, however, is that no gender differences emerged with regard to the Judge Trial, which also has an adversarial nature.

An implication of the lack of gender differences in this category is that women find an exception for Judge Trials compared to these other forms of adjudication. One possible explanation is that women might feel that judges convey authority or deserve respect in a way that arbitrators and jurors do not. If that authority appeals to females as much as males, there would be no gender difference for attraction to the Judge Trial. An analysis performed to explore this interpretation supported this possibility.148 It revealed that men tended to give the Jury Trial and Binding Arbitration—but not the Judge Trial—significantly higher ratings than did women.149

This analysis also revealed that both men150 and women151 liked the Judge Trial more than both the Jury Trial and Binding Arbitration. However, men found the Jury Trial and Binding Arbitration to be equivocal in terms of attractiveness,152 whereas women liked the Jury Trial significantly more than Binding Arbitration.153 Together, these findings suggest that, although both men and women  liked the Judge Trial the most, women were more resistant to Binding Arbitration compared to men, and preferred to use either the Judge or Jury Trial—but especially the Judge Trial—to resolve their dispute.

4.     Personal Injury Litigants Liked the Jury Trial More than Property Litigants, but Case Type Was Not a Major Predictor Otherwise

To determine the effect of case type as a predictor of attraction to each procedure, personal injury (the most common case type in the sample) was used as the reference group. The data revealed that those whose cases concerned only personal injury matters liked the Jury Trial significantly more than those whose cases involved only property issues.154

This finding resonates with the widely held perception that jury sympathy in personal injury cases results in high damage awards to plaintiffs.155 Yet, the appeal of the Jury Trial was not higher for plaintiffs versus defendants. This pattern is curious until one considers that litigants’ attraction to the Jury Trial was found to be related to their confidence in their case (i.e., higher confidence was associated with greater attraction to the Jury Trial, and vice versa). Thus, it is possible that plaintiffs and defendants in personal injury cases were equally attracted to the Jury Trial, but that their confidence in a trial win better explained attraction to this procedure. A follow-up analysis designed to test this possibility revealed that the relation between attraction to the Jury Trial and confidence in a trial win for personal injury litigants did not differ significantly between plaintiffs and defendants.156 This result supports the idea that litigants’ attraction to the Jury Trial in personal injury cases was better explained by the confidence they had in their case than by their status as either a plaintiff or defendant.

Case type mattered in relatively few other instances. Personal injury litigants liked the Attorneys Negotiate with Clients Present option less than those with “other” kinds of cases. They also liked the Judge Decides without Trial option and the Judge Trial significantly less than those with multiple case types. The latter finding suggests that those with more legally complicated disputes valued the prospect of having a judge decide their case more than did those whose cases concerned personal injury matters only.

5.     Relationship Variables Were Associated with Attraction to the Negotiation Options, but Not with Attraction to Adversarial Procedures such as Binding Arbitration or Trials

An interesting pattern emerged regarding the relationship between the parties and how they perceived their negotiation options.157 If litigants had a pre-existing relationship with the opposing party, they liked Attorneys Negotiate without Clients less, and vice versa. But litigants with a pre-existing relationship did not differ from those without one in terms of how much they liked the Attorneys Negotiate with Clients Present option. This somewhat counterintuitive pattern suggests that although litigants who had a relationship with the opposing party were agnostic about the option that would allow them to interact with the other party, something about negotiations taking place without them was relatively unappealing.

By contrast, the more that litigants valued a future relationship with the other party, the more they liked Attorneys Negotiate with the Clients Present, and vice versa. This finding makes intuitive sense. If litigants desire a future relationship, it seems wise to informally work together to resolve the conflict; if a future relationship is not important, efforts to collaborate might be less appealing.

Although one might intuit that litigants who highly value a future relationship might be less attracted to adjudicative or adversarial procedures—specifically, Judge Decides without Trial, the Jury Trial, the Judge Trial, and Binding Arbitration—compared to those less interested in a future relationship, the data did not support this notion. This pattern might shed light on litigants’ resistance vis-à-vis more collaborative models when they regard relationship preservation as a priority. Specifically, litigants might not anticipate the adverse effects that more adversarial procedures might have on relationships,158 or they might believe that the benefits to having a neutral third-party decide their case outweigh any negative consequences.

6.     Court Impressions Related to Attraction to Judicial Decisions

The data revealed that the more favorably the litigants rated the court where their case was filed, the more they liked the two options that granted decision-making control to a judge—namely, Judge Decides without Trial and the Judge Trial. The less favorably they viewed the court, the less they liked these two options. This pattern resonates with research suggesting that greater perceived institutional legitimacy is associated with a stronger preference for, and acceptance of, court decisions.159 The only other procedure that was significantly associated with their impressions of the court was Attorneys Negotiate without Clients. The more litigants liked the court, the more they liked this procedure and vice versa.

E.     Demographic Variables

Surprisingly, factors that previous scholars have speculated or observed to be associated with procedural preferences were rarely, if ever, found to be significant predictors of attraction to procedures in the present study. For example, gender was associated with an attraction to the Jury Trial and Binding Arbitration only.160 Ethnicity was not a factor in any of the significant models.161 Age group was significantly related only to attraction to Attorneys Negotiate with Clients Present162 (i.e., those in older age groups liked this option less than those in younger ones, and vice versa). Role in the case was significant only for Binding Arbitration (i.e., litigants acting as both a plaintiff and defendant liked Binding Arbitration more than those acting only as plaintiffs). Court location was a significant predictor only for the Jury Trial (i.e., those with cases in California liked the Jury Trial less than those with cases filed in Oregon). The interest of an insurance company in the outcome of the dispute was statistically significant only for Attorneys Negotiate with Clients Present (i.e., if an insurance company was involved, this option was more attractive compared to cases which did not involve an insurance company). Party type (i.e., whether the litigant was an individual, a company, group or organization) mattered only for the two types of negotiation (i.e., companies liked the option of Attorneys Negotiate without Clients more than individual litigants did, and litigants representing a group or organization liked the Attorneys Negotiate with Clients Present option less than individual litigants did). Opposing party-type was relevant for predicting attraction to the Jury Trial and Binding Arbitration only (i.e., compared to litigants who opposed individuals, those opposing groups or organizations liked the Jury Trial more and those opposing companies liked Binding Arbitration more).

Thus, some of the demographic, case type, relationship, and attitudinal factors found to be predictive of attraction to procedures in past research were not found to be significant in the present study.163 One possible explanation is that whereas previous research generally examined only one or two specific predictors within a given study,164 the present study evaluated the relation of many of these previously disparate predictors while controlling for their shared effects as predictors. While no model is perfect, such multivariate regression models can help to differentiate between variables previously thought to be associated with attraction for procedures that in fact merely share a fraction of association with some other, better predictor. Future research might identify other factors, or interactions between factors, that could explain even greater variability in attractiveness ratings, such as the amount in controversy, whether litigants’ previous experiences with the legal system were positive or negative, or their socio-economic status.

Even though previous studies found certain factors to be predictive when evaluated individually, the overall pattern suggests that when a multitude of factors are considered simultaneously, relatively few may actually be associated with attraction for procedures. This take-away is likely to come as a surprise to lawyers or court administrators who have strong views regarding which procedure is likely to appeal to “a certain kind of litigant” or “someone with a certain kind of case.”

F.     Caveats Regarding Interpretations of the Findings

This study provides considerable new information about litigants’ ex ante procedural preferences.  But, as with any empirical study, one should keep a few characteristics of the methodology in mind when relying on the findings.

First, and most importantly, the data from this study should not be used to infer causal relationships between the factors that were examined. Causal relationships can be properly ascertained only through controlled laboratory experiments.165 Analyses based on the current dataset are helpful for understanding relationships between variables such as attraction to procedures and their expected use of those procedures, but not for deducing whether a particular causal relationship exists (i.e., whether expectation of use leads to attraction, or vice versa). Thus, although the interpretations of analyses that are explored in this Article are consistent with the analyses that were reported, they should not be viewed as evidence that particular causal relationships were found.

One possible limitation on the generalizability of the findings concerns the response rate. Although a higher response rate would have been preferable, and would reduce the possibility that certain kinds of litigants or cases were underrepresented in the sample, a 10% response rate is not unusual for a survey of laypeople who are solicited through the mail.166 Moreover, the 10% response rate is higher than that of the other ex ante field studies on litigants’ procedural preferences.167

A number of factors could have limited the response rate. First, given that the survey asked questions about an active dispute, some litigants may have withheld responses out of privacy or confidentiality concerns. Second, the survey length—nine pages containing forty-nine questions—may have deterred participation.168Third, because we typically needed to research disputants’ addresses, it is possible that some of the addresses we used were incorrect. In such instances, we would not expect a returned survey for our efforts. In this light, the calculated response rate is likely to be a gross underestimate of the true response rate and may greatly understate the representativeness of the sample. Replications of this research might implement strategies to obtain a higher response rate, such as greater compensation or asking the court to collect address information for each litigant.

Although the present study represents a substantial improvement over past research in terms of breadth, it analyzes cases filed in only three courts. To examine the generalizability of the findings, future investigations might attempt to replicate the study in other jurisdictions.169 It is worth noting, however, that the dataset ultimately included litigants whose mailing addresses were located in nineteen different states. In this respect, the sample did represent some geographical diversity.

Another possible limitation to generalizability is that the sample included significantly more plaintiffs than defendants.170 This outcome is unexpected given that relatively more surveys were mailed to defendants than plaintiffs (i.e., the selection procedure used to generate the sample suggests that there were relatively more defendants than plaintiffs in the cases filed during the recruitment period).171 Thus, all else being equal, the opposite finding regarding returned surveys should have been observed. But obviously all was not equal. Plaintiffs may have been more inclined to discuss their cases than defendants. After all, by virtue of the fact that they were the initiators of the lawsuit, they were at least somewhat eager to publicly discuss the conflict. By contrast, defendants may have been comparatively more concerned about privacy and confidentiality and thus withheld participation at higher rates.

A final caveat regarding generalizability concerns the kind of cases that were excluded from the study. The sample included only civil cases for which the study courts offered both mediation and arbitration in addition to trial. Foreclosure and collections cases were ineligible on the grounds that individuals on the corporate side who might be familiar enough with the case to have been recruited for the study (i.e., managers, collections officers) oversee so many similar cases that their recollection of case details would be unreliable. Landlord–tenant, family law, tax, and bankruptcy cases were ineligible because, in many jurisdictions, they belong to specialized courts.172 The decision to exclude such cases served to maximize the generalizability of the data to courts of general jurisdiction. It is possible that procedural preferences for cases normally filed in specialized courts might differ from the preferences elucidated in the present study.

VI.     Conclusion

This Article began with a quote by Tom Tyler. He has argued that legal authorities can simultaneously do well at their jobs and create public satisfaction, provided that they gain a clear understanding of what laypeople want from the courts, which in turn requires examining their preferences concerning how disputes should be resolved.173

Not only can legal authorities bridge public satisfaction with a job well done, they must; and knowledge of litigant preferences is paramount to the achievement of that goal. The study presented here sets the stage for legal professionals to develop protocols, procedures, and best practices that serve the goals of the legal system as well as the expectations of the litigants who seek justice.

An important take-away from the study is that litigants do indeed have procedural preferences. None of the procedures were rated significantly higher in attractiveness than the Judge Trial, Mediation, or Attorneys Negotiate with Clients Present. These preferences help to clarify the ways in which findings elucidated from laboratory research generalize, or fail to generalize, to the contemporary legal world.

Results from the present research can inform legal actors such as lawyers, court administrators, and judges about litigants’ general perceptions of procedures and how their demographics, case types, relationships and attitudes relate, and do not relate, to how they perceive their options. Legal actors can rely on the findings to anticipate preferences in favor of some procedures, and relative resistance against the use of others, which may vary given how much the litigants value a future relationship with the other party, how they perceive the court where their case was filed, or how confident they are in their case. To the extent that litigants’ perceptions of procedures differ from those of lawyers, some differences might be due to litigants’ misconceptions about those procedures, whereas others might reflect incorrect assumptions that lawyers have about what litigants value or hope to accomplish through civil litigation. Research that elucidates the litigant’s perspective can foster collaboration between legal professionals and litigants and enrich their relationships.

Ideally, future research will fill the gaps in the existing literature in ways that will serve both lawyering and court policy. But, in the end, the advancement of procedural justice in light of litigants’ preferences will depend on legal actors doing their part to implement such research. By catalyzing empirical research findings into policy, courts can further the goals of democratic governance, and foster respect for, and trust of, the civil justice system.

Acknowledgements

This material is based upon work supported by the National Science Foundation under Grant Number 0920995. The American Bar Association, Section on Litigation, and the Institute for Governmental Affairs at the University of California, Davis provided financial support during the first year of the study. The University of California, Davis, School of Law provided ongoing financial and human resources to ensure the success of this project. Special gratitude is extended to Judge Beeman and Grace Andres from the Solano Superior Court in California, Judge Kristena A. LaMar and Doug Bray from the Multnomah County Circuit Court in Oregon, and Nini Rich from the Utah State Courts system for their indispensable enthusiasm for this project and for facilitating the participation of the courts within their jurisdictions.

Special thanks are extended to Laura Castro-Schilo and Joseph Gonzales for the significant statistical expertise they contributed to this project.This project also greatly benefited from feedback from Erin Archerd, Robert Artuz, Lisa Blomgren Bingham, Jeanne Brett, Deborah Goldfarb, Deborah Hensler, David Horton, Randall Kiser, John Lande, Jan Martinez, Barbara “Bobbi” McAdoo, Miguel Mendez, Danny Oppenheimer, Lee Ross, Donna Steinstra, and Nancy Welsh. Great appreciation is extended to research assistants Rebekah Barlow, Caitlin Barth, Mitesh Bhatka, Barbara Borkowski, Marvin Cho, Shannon Clawson, Lindsay Gold, David Jefferson, Jihan Kahssay, Josephine Lee, Erin Mariano, Sabrina Muerle, Hovannes Nalbandyan, Sean Newland, Jan Roos, Saba Shatara, Tiffany Tran, Melanie Young, and to law librarians Susan Ilano, Elisabeth McKechnie, and Erin Murphy. Linda Cooper, Jennifer Angeles, Lori Dana, Christine Hernandez, Elisabeth Leisure, Shelagh MacKay, and Sue Williams provided indispensable administrative support. Gratitude is extended to Anne Skove, Richard Shauffler, and the National Center for State Courts for their assistance in locating potential study courts, and to Nancy Welsh and Joe Barrett from the U.S. District Court for the Middle District of Pennsylvania for facilitating access to pilot subjects. This Article also benefited from the Author’s presentations at Stanford Law School, Florida State University College of Law, and the Northern California ADR Faculty Conference, as well as conferences sponsored by the American Bar Association Section on ADR, , the California Dispute Resolution Council, and the American Psychology-Law Association.

Gratitude is also extended to Jeanne Brett, Deborah Hensler, and Leonard Horowitz whose past mentorship undoubtedly inspired the design and execution of this study. Any errors, of course, belong solely to the Author.

Appendices

Appendix A: Eligible Case Types by Court

Shest 11

Note: In rows where there is more than one case type (separated by “/”), the following key indicates which specific eligible case types are associated with particular courts: Bold = Utah Court; Italic = California Court; Underline = Oregon Court. Some case types are eligible in more than one court. For example, “Property Damage” means that for both the Utah Court and the Oregon Court this was an eligible case type.

Appendix B: Study Courts

The Utah Court. In accordance with Utah’s Alternative Dispute Resolution Act,174 the Utah Court’s ADR program applies to all civil cases filed after January 1, 1995.175 Litigants may effectively choose from among mediation, non-binding arbitration, and trial.176 Although certain narrow types of cases are excluded from ADR by statute,177 claims are not referred to different ADR procedures on the basis of the amount in controversy or type of case.178

Under the court’s procedures, any civil matter subject to Rule 4-510 will proceed to mediation within thirty days after the filing of the responsive pleading.179 However, litigants may opt out of mediation and into non-binding arbitration180 or binding arbitration181 by filing a written agreement signed by counsel and the parties before the thirty days after the filing of the responsive pleading have passed. Parties using arbitration or mediation may select a neutral from the Court-Annexed ADR Provider Roster.182 Parties may proceed to trial, even once the matter has been referred to ADR, by making a motion to the court.183 The court also retains the traditional right to dismiss the case.184

The California Court. In accordance with California’s Dispute Resolution Programs Act, the California Court encourages litigants to utilize alternatives to trial.185 They have the option of non-binding arbitration, mediation, or trial for most disputes. As in the Utah Court, ADR is not mandatory in the California Court186 and all civil cases not barred from ADR by statute are eligible for the court’s own mediation and non-binding arbitration programs regardless of the amount in controversy.187

Unlike the Utah Court, the California Court’s litigants are not automatically assigned to an ADR procedure.188 Rather than go through the additional step of petitioning to opt out of an ADR procedure in favor of trial, as in the Utah Court, litigants “opt in” from trial to mediation or arbitration.189 Parties may divert190 the dispute to mediation or arbitration once the court sends “a flyer explaining the alternatives to resolve [the] dispute, the different types of ADR, and a stipulation and order to ADR.”191 These materials are sent to the plaintiff’s attorney of record or the pro per plaintiff upon the filing of the complaint and to the defendant upon filing the answer.192 Alternatives to trial are revisited at the first case management conference.193 At this time, the parties may inform the judge if they are willing to participate in any of the offered programs.194 If referred to mediation at the case management conference, parties must notify the court of their mutual choice, or individual selection, of a mediator within twenty days of referral to mediation.195 Time limits for entering into mediation or arbitration are set by the judge when the case is referred. 196

The Oregon Court. Like all circuit courts in Oregon, the Oregon Court is subject to a state statute mandating non-binding arbitration197 when the amount in controversy does not exceed $50,000.198 Civil cases with any claim for an amount exceeding $50,000, or cases seeking specific relief and non-monetary claims, are not subject to the mandatory arbitration provision.199 However, parties to any case not subject to mandatory arbitration may use the ADR programs on a voluntary basis and may select either mediation or arbitration upon stipulation to the court.200

Parties wishing to avoid mandatory arbitration have a number of options. First, a recently enacted uniform court rule implemented by the Chief Justice allows parties to stipulate to a jury trial if the parties seek to categorize the case as an expedited one.201 If the presiding judge accepts the stipulation, a case may be removed from mandatory arbitration and set for a court date “no later than four months from the date of the order” granting expedited status.202

Second, similar to the Utah Court, litigants wishing to avoid mandatory arbitration may file a “Motion for Exemption from Arbitration.”203 Upon such a motion, the judge presiding over the case has discretion to remove it from mandatory arbitration and send the matter to trial.204 While no judge (as of August 2012) has exempted from referral any classes of cases subject to the mandatory arbitration statute, individual cases are routinely removed from arbitration for good cause shown.205 In addition, litigants referred to arbitration have the option to proceed to mediation rather than arbitration on stipulation filed with the circuit court prior to the arbitrator commencing the arbitration hearing.206 If mediation is stipulated in lieu of arbitration, the mediation must be completed within the same time frame as arbitration.207

Subject to narrow exceptions, the court also mandates participation in some form of ADR within 270 days from the filing of the first complaint.208 Under Multnomah County Court Rule 7.075, parties may choose, but are not limited, to engage in arbitration, mediation, or judicial settlement conferences.209

There is no way for litigants to opt out of Rule 7.075, but its application is relatively limited given the options they have to avoid mandatory arbitration. Aside from the nature of cases it addresses, the limited use of Rule 7.075 is also due in part to the court’s master calendar management of civil actions.210 Often, the presiding judge will designate a case as complex under Oregon Uniform Trial Court Rule 7.030, thereby avoiding application of Rule 7.075.211

Cases set for trial will not be assigned to arbitration within sixty-three days of the trial date without a court order, unless all parties agree on an arbitrator and hearing date at least twenty-eight days before trial.212 Parties may opt for mediation once a matter has been set for mandatory arbitration at any time prior to the arbitration hearing by stipulating to the mediation in writing.213 A case removed from arbitration to mediation must complete the mediation in the same time period permitted for arbitration.214

Appendix C: Participant Recruitment

Law student research assistants assisted with participant recruitment. For the litigants whose addresses were not provided by the court, they conducted searches using a combination of LexisNexis (Locate a Person Nationwide, Locate a Business Nationwide, Corporation Filings, Real Property Assets, Professional Licenses, Voter Registration, Jigsaw People Finder, and Zoom People Information), Dogpile, Zabasearch, Spokeo, Martindale, Facebook, Switchboard, State Bar websites, and Google. Because the Oregon and California Courts did not routinely collect litigant contact information (except, for example, when a litigant was pro per), they researched contact information in the manner described above. They also relied on information from the case file, when available, to narrow down the searches. For example, if the file reported a litigant’s middle initial, that information was used pinpoint a person with a common name. If spouses were named as parties to the lawsuit, or the litigant’s profession could be deduced, this information also informed the searches.

When the litigant was a collective (i.e., company group or organization), the address research took one of several routes. For small collectives, the team used the Internet to identify the owner or president, and then followed the aforementioned search method. For large, well-known collectives whose lead attorney was identified in the case file, the team found the attorney’s contact information using the same method and treated the attorney as the representative litigant. When the lead attorney was not identified in the case file, Internet research was used to find the name of the head of that collective’s legal division in the state where the case was filed, and the team treated that individual as the representative litigant. On rare occasions, the team called the company to ascertain the relevant information. It was more common for attorney contact information to be provided for plaintiffs compared to defendants. When the research assistants exhausted all research strategies and could not obtain contact information for a litigant, they awarded a certainty score of “1.”

Appendix D: Descriptions of Procedures

How attractive do you find each alternative for your particular case? Please circle the appropriate number for each option.

Attorneys Negotiate without Clients: The lawyers negotiate with each other on behalf of their clients, in order to settle their client’s case. The negotiations may be done in person (face-to-face), but are more often done by phone, fax, or email. The clients are never present at any of the negotiation discussions. The outcome could be based on the law, or it could be based on some other rules or principles that the clients and/or their lawyers find relevant or important. If the lawyers agree on an outcome that their clients find acceptable, that is the outcome for the case.

Shestowksy 1Attorneys Negotiate with Clients Present: The lawyers negotiate with each other on behalf of their clients, in order to settle their clients’ case. The negotiations may be done in person (face-to-face) with lawyers and clients present, or by phone, with lawyers and clients on the phone at the same time. The clients are present and may participate in the negotiation discussions. The outcome could be based on the law, or it could be based on some other rules or principles that the clients and/or their lawyers find relevant or important. If the clients agree on an outcome that they both find acceptable, that is the outcome for the case.

Shestowksy 3Mediation: A mediator (a neutral third person) facilitates the discussion between opposing lawyers and clients to help them settle the case. The mediator has no power to decide the outcome. Instead, the mediator helps the lawyers and clients communicate their different perspectives, discuss their needs and interests, and explore ways to resolve the case in a way that is acceptable to both clients. The outcome could be based on the law, or it could be based on some other rules or principles that the clients and/or their lawyers find relevant or important. If the clients agree on an outcome that they both find acceptable, that is the outcome for the case.

Shestowksy 4Judge Decides without Trial: Sometimes a judge can decide a case early on, so that a trial is never required. This is because the judge has determined there is no question about the facts, and the case can be decided on the basis of law alone. The lawyers submit documents to the court and may make a presentation to the judge at a hearing. Clients rarely attend and, if they do, they do not speak during the hearing. The judge later announces the outcome in writing, and explains why they decided as they did. This outcome is based on legal rules or principles. A party who is dissatisfied with the outcome can appeal it to a higher court, which will require additional time and proceedings.

Shestowksy 5Jury Trial: A jury hears the lawyers’ arguments and witnesses’ answers to the lawyers’ questions and reviews physical evidence. Often, the clients testify. The plaintiff’s (party making the claim) lawyer presents his/her case first and then the defendant’s lawyer presents his/her case. The judge instructs the jurors about the law they must apply to the evidence they have seen and heard. After the evidence has been presented, the jurors decide the outcome in private and then announce it in court, without explaining why they decided as they did. This outcome is based on legal rules or principles. A party who is dissatisfied with the outcome can appeal it to a higher court, which will require additional time and proceedings.

Shestowksy 6Judge Trial: A judge hears the lawyers’ arguments and witnesses’ answers to the lawyers’ questions and reviews physical evidence. Often, the clients testify. The plaintiff’s (party making the claim) lawyer presents his/her case first and then the defendant’s lawyer presents his/her case. After the evidence has been presented, the judge decides the outcome in private and announces it in writing, and explains why they decided as they did. This outcome is based on legal rules or principles. A party who is dissatisfied with the outcome can appeal it to a higher court, which will require additional time and proceedings.

Shestowksy 7Binding Arbitration: The clients and lawyers select an arbitrator (or group of arbitrators). Arbitrators are neutral third parties who are not actual judges, but who will have the authority to decide the outcome of the case. The clients will also decide among themselves what the rules for presenting evidence to the arbitrator will be. These rules are generally less formal than the rules used at trial. The lawyers, their clients, and client’s witnesses present evidence. Often, the clients testify. After the evidence has been presented, the arbitrator(s) decide(s) the outcome of the case. This outcome is typically based on legal rules or principles, but it could be based on other rules or principles that the clients agree in advance should be used to decide the outcome, such as industry standards. Arbitrators typically report the outcome in writing, without explaining why they decided as they did. Because this type of arbitration is “binding,” it is extremely difficult to appeal their decisions.

Shestowksy 8Non-binding Arbitration: The clients and lawyers select an arbitrator (or group of arbitrators). Arbitrators are neutral third parties who are not actual judges, but who will have the authority to decide the outcome of the case. The clients will also decide among themselves what the rules for presenting evidence to the arbitrator will be. These rules are generally less formal than the rules used at trial. The lawyers, their clients, and client’s witnesses present evidence. Often, the clients testify. After the evidence has been presented, the arbitrator(s) decide(s) the outcome of the case. This outcome is typically based on legal rules or principles, but it could be based on other rules or principles that the clients agree in advance should be used to decide the outcome, such as industry standards. Arbitrators typically report the outcome in writing, without explaining the reasons why they decided as they did. Afterwards, because this type of arbitration is “non-binding,” either of the clients may reject the outcome and go to trial.

Shestowksy 9Appendix E: Descriptive Statistics of Untransformed Continuous Variables from Regression Model

Shest 12

Note: The range of acceptable skewness scores was +/- 2, and the range of acceptable kurtosis scores was +/- 3, with scores unbounded by +/- 2 considered moderate. Variables with kurtosis scores beyond the range of +/- 2 were transformed using root, power, and log functions to improve normality of variable distribution.

Appendix F: Factors Predicting the Attractiveness of Attorneys Negotiate Without Clients

Shest 13

** p < .01. * p < .05. † p < .10. Note: B indicates the unstandardized coefficients, SE indicates unstandardized coefficients’ standard errors, t indicates t-test values for the unstandardized coefficients, p indicates the corresponding probability values for the unstandardized coefficient t-test values, and β indicates the standardized coefficients.

Appendix G: Factors Predicting the Attractiveness of Attorneys Negotiate with Clients Present

Shest 14

** p < .01. * p < .05. † p < .10.

Appendix H: Factors Predicting the Attractiveness of the Jury Trial

Shest 15

* p < .05. † p < .10.

Appendix I: Factors Predicting the Attractiveness of the Judge Trial

Shest 16

** p < .01. * p < .05. † p < .10.

Appendix J: Factors Predicting the Attractiveness of Judge Decides Without Trial

Shest 17

** p < .01. * p < .05. † p < .10.

Appendix K: Factors Predicting the Attractiveness of Binding Arbitration

Shest 18

** p < .01. * p < .05. † p < .10. 


  1. [1]. See Sara Murray, Slump Over, Pain Persists, Wall St. J. (Sept. 21, 2010, 12:01 AM), http://online.wsj.com/article/SB10001424052748703989304575503691644231892.html.

  2. [2]. As States Cut Court Budgets, Who Pays the Price?, NPR (Oct. 4, 2011, 12:00 PM), http://www.npr.org/templates/story/story.php?storyId=141043681 (interviewing Mary McQueen, president of the National Center for State Courts).

  3. [3]. Id. However, the number of dispositions continues to grow despite the decline in the number of trials. See Brian J. Ostrom et al., Examining Trial Trends in State Courts: 1976–2002, 1 J. Empirical Legal Stud. 755, 773 (2004).

  4. [4]. See Stephen Stock, California Superior Courts in Crisis, NBC Bay Area (July 24, 2013, 10:05 AM), http://www.nbcbayarea.com/news/local/California-Superior-Courts-in-Crisis-2166
    68081.html.

  5. [5]. For example, in 2011, budgets for some New York ADR centers were reduced by nearly 40%. Matt Chandler, Campaign to Restore Child Mediation Funds Launched, Buffalo Bus. First (Jan. 11, 2012, 11:11 AM), http://www.bizjournals.com/buffalo/news/2012/01/11/campaign-to-restore-childmediation.html?ana=RSS&s=article_search&utm_source=feedburner&utm_medium=feed&
    utm_campaign=Feed:+bizj_buffalo+%28Business+First+of+Buffalo%29. In Florida, the governor’s office siphoned $14 million from trust funds that had been earmarked, in part, for mediation and arbitration programs, and, in 2011, “North Carolina completely eliminated state funding for court ADR.” Heather Scheiwe Kulp, Top Ten Court ADR Developments in 2011, Trends for 2012, Resol. Systems Inst. (Jan. 10, 2012), http://blog.aboutrsi.org/2012/program-evaluation/top-ten-court-adr-developments-in-2011-trends-for-2012/.

  6. [6]. Donna Shestowsky, Disputants’ Preferences for Court-Connected Dispute Resolution Procedures: Why We Should Care and Why We Know So Little, 23 Ohio St. J. on Disp. Resol. 549, 587 (2008).

  7. [7]. It is within the power of many, if not most, courts to design dispute resolution programs that comport with litigants’ preferences for procedures if the courts choose to do so. “Although legislation in some jurisdictions might require parties to use ADR for certain types of cases, program design is largely determined by the courts. . . .” Id. at 583–84 (citing Caroline Harris Crowne, Note, The Alternative Dispute Resolution Act of 1998: Implementing a New Paradigm of Justice, 76 N.Y.U. L. Rev. 1768, 1794–95 (2001)). Court programs vary in a variety of ways, including the number of procedures they offer, the type of procedures they offer, and whether the litigants get to choose the procedure or whether one is mandated by the court. See id. at 584–85. For an analysis of disputants’ ability to “self-determine” dispute resolution procedures, see Lisa B. Bingham, Why Suppose? Let’s Find Out: A Public Policy Research Program on Dispute Resolution, 2002 J. Disp. Resol. 101.

  8. [8]. Thomas J. Stipanowich, ADR and the “Vanishing Trial”: The Growth and Impact of “Alternative Dispute Resolution,1 J. Empirical Legal Stud. 843, 843 (2004) (arguing that where there has been “evidence of cost and time savings . . . [in] some court-[connected] ADR programs, it is evident that much depends on the shape and structure of such programs”).

  9. [9]. Although understanding ex post evaluations of procedures is important for many reasons, including understanding how litigants feel about their experiences with the legal system, and how likely they are to voluntarily comply with the outcomes of the procedures, this Article’s focus is on the ex ante perspective.

  10. [10]. For a thoughtful review of how the term “procedural justice” has been defined, see Tom R. Tyler, Why People Obey the Law 3–7 (1990). In this Article, the term refers to the perceived fairness of the procedures used to resolve disputes or allocate resources.

  11. [11]. For a review of this research, see Shestowsky, supra note 6.

  12. [12]. See Robert J. MacCoun, Voice, Control, and Belonging: The Double-Edged Sword of Procedural Fairness, 1 Ann. Rev. L. & Soc. Sci. 171, 173–74 (2005).

  13. [13]. See Donna Shestowsky & Jeanne Brett, Disputants’ Perceptions of Dispute Resolution Procedures: An Ex Ante and Ex Post Longitudinal Empirical Study, 41 Conn. L. Rev. 63, n.46 (2008) (employing a longitudinal design to compare the ex ante and ex post evaluations of disputants involved in legal procedures and concluding that just one past study examined disputants’ ex ante evaluations); Lamont E. Stallworth & Linda K. Stroh, WHO is Seeking to Use ADR? WHY Do They Choose to Do So?, Disp. Resol. J., Jan.–Mar. 1996, at 30, 33–35 (using data from disputants with pending disputes to explore ex ante attitudes about procedures).

  14. [14]. In this Article, reference to “our” and “we” refer to the research team members at the University of California, Davis, School of Law. The team was composed primarily of law students whom the Author selected, trained, and supervised. Any errors are the sole responsibility of the Author.

  15. [15]. This assertion is based on a thorough review of research published in legal and social science journals as of February 2013.

  16. [16]. See Tyler, supra note 10, at 3–5 (describing previous literature). See generally Donna Shestowsky, Procedural Preferences in Alternative Dispute Resolution: A Closer, Modern Look at an Old Idea, 10 Psychol. Pub. Pol’y & L. 211, 216–18 (2004) (reviewing the relevant literature).

  17. [17]. See Gerald F. Phillips, The Obligation of Attorneys to Inform Clients About ADR, 31 W. St. U. L. Rev. 239 (2004). For example, the Sacramento County Bar Association mandates that “[l]awyers shall . . . [a]dvise the client at the outset of the availability of alternative dispute resolution and explain in simple language what the effects of the various ADR techniques, e.g., mediation, neutral evaluation or mini-trial might have on the case.” Standards of Professional Conduct: Section 8: Settlement and Alternative Dispute Resolution, Sacremento Cnty. Bar Ass’n, http://www.sacbar.org/About%20SCBA/Documents/professionalconduct.aspx (last visited Nov. 17, 2013). Similarly, the Virginia Supreme Court’s rules provide that “The client has ultimate authority to determine the purposes to be served by legal representation . . . . In that context, a lawyer shall advise the client about the advantages, disadvantages, and availability of dispute resolution processes that might be appropriate in pursuing these objectives.” Va. Rules of Prof’l Conduct R. 1.2 cmt. 1 (West 2013).

  18. [18]. For a discussion of how lawyer and client perceptions about cases and procedures often differ, and how lawyers’ incorrect assumptions can hinder their understanding of the client’s perspective, see Phillips, supra note 17, at 251–52 (discussing the possibility that attorneys and their clients may differ in how they want to handle a case); Shestowsky, supra note 6, at 593–98.

  19. [19]. Attorneys’ perceptions of their clients’ litigation goals often diverge from their actual goals. For example, attorneys often fail to appreciate the importance of noneconomic factors to plaintiffs, which could include “objectives of obtaining admissions of fault, acknowledgments of harm, retribution for defendant conduct, prevention of reoccurrences, answers, and apologies.” Tamara Relis, Perceptions in Litigation and Mediation: Lawyers, Defendants, Plaintiffs, and Gendered Parties 34 (2009).

  20. [20]. Edgardo Buscaglia & Paul B. Stephan, An Empirical Assessment of the Impact of Formal Versus Informal Dispute Resolution on Poverty: A Governance-Based Approach, 25 Int’l Rev. L. & Econ. 89, 90 (2005).

  21. [21]. See, e.g., John Lande, Using Dispute System Design Methods to Promote Good-Faith Participation in Court-Connected Mediation Programs, 50 UCLA L. Rev. 69, 118–20 (2002) (discussing how mediation program planners can incorporate empirical findings concerning disputants’ subjective perceptions to promote productive participation in mediation, in order to solve problems associated with apparent bad faith conduct).

  22. [22]. Tom R. Tyler, Trust and Law Abidingness: A Proactive Model of Social Regulation, 81 B.U. L. Rev. 361, 391 (2001); see also Wayne D. Brazil, Early Neutral Evaluation 21–22 (2012); Susan M. Olson & David A. Huth, Explaining Public Attitudes Toward Local Courts, 20 Just. Sys. J. 41, 41–42 (1998) (explaining the need for the majority of citizens to respect court decisions and comply with court orders voluntarily).

  23. [23]. See Murray, supra note 1 (explaining the economic hardships caused by the 2008 recession).

  24. [24]. See, e.g., Craig A. McEwen & Richard J. Maiman, Mediation in Small Claims Court: Achieving Compliance Through Consent, 18 Law & Soc’y Rev. 11, 20–22 (1984) (concluding that litigants in consensual procedures such as mediation are more likely to perceive the outcome as fair and just and, subsequently, are more likely to comply with the outcome than in adjudicated cases); Mark S. Umbreit et al., Victim-Offender Mediation: Three Decades of Practice and Research, 22 Conflict Resol. Q. 279, 298–99 (2004) (concluding that offenders who participate in programs that offer them greater opportunity to shape the outcome are more likely to comply with the outcome and are less likely to re-offend than those who engage in procedures that are more adjudicative).

  25. [25]. See Tom R. Tyler, The Psychology of Disputant Concerns in Mediation, 3 Negotiation J. 367, 368 (1987); Umbreit et al., supra note 24, at 298. As Tom Tyler has argued, on the basis of compelling empirical research, procedures that subjectively appeal to litigants can inspire people to “obey the law” and reduce the need for governmental intervention to ensure legal compliance. See Tyler, supra note 10.

  26. [26]. For a review, see Shestowsky & Brett, supra note 13, at 68–73.

  27. [27]. For a review of the relevant literature, see Shestowsky, supra note 6; Shestowsky, supra note 16.

  28. [28]. For a review of early laboratory research, see Jeffrey Z. Rubin, Experimental Research on Third-Party Intervention in Conflict: Toward Some Generalizations, 87 Psychol. Bull. 379 (1980).

  29. [29]. See, e.g., William Austin et al., Effect of Mode of Adjudication, Presence of Defense Counsel, and Favorability of Verdict on Observers’ Evaluation of a Criminal Trial, 11 J. Applied. Soc. Psychol. 281, 291 (1981); Stephen LaTour, Determinants of Participant and Observer Satisfaction with Adversary and Inquisitorial Modes of Adjudication, 36 J. Personality & Soc. Psychol. 1531, 1535–36 (1978); Laurens Walker et al., Reactions of Participants and Observers to Modes of Adjudication, 4 J. Applied. Soc. Psychol. 295, 300 (1974).

  30. [30]. See Shestowsky, supra note 6, at 606.

  31. [31]. Scholars have often raised concerns regarding the ecological validity of social science research in law. Brian H. Bornstein, The Ecological Validity of Jury Simulations: Is the Jury Still Out?, 23 Law & Hum. Behav. 75, 75–76 (1999) (explaining that common validity concerns with laboratory research include (1) the reliance on undergraduates for data (i.e., the concern that the preferences of undergraduates and community-dwelling adults differ), (2) “the research setting” (i.e., laboratory vs. courtroom), and (3) “the consequentiality of the task” (i.e., making a hypothetical decision vs. a decision with real-life consequences)); see also Wayne Weiten & Shari Seidman Diamond, A Critical Review of the Jury Simulation Paradigm: The Case of Defendant Characteristics, 3 Law & Hum. Behav. 71, 75–83 (1979) (identifying problems in the external validity of laboratory studies, including “inadequate sampling,” “inappropriate dependent variables,” “lack of corroborative field data,” and the differences in consequences between real versus simulated situations). Yet, despite concerns over the ecological validity of laboratory research, some evidence suggests that such trepidation is not warranted. See MacCoun, supra note 12. Reviews of field research have also been replete with criticisms for having limited utility for drawing conclusions about causation, and for collecting data from a single (and perhaps aberrant) court. See Bobbi McAdoo et al., Institutionalization: What Do Empirical Studies Tell Us About Court Mediation?, Disp. Resol. Mag., Winter 2003, at 8, 8.

  32. [32]. See, e.g., E. Allan Lind et al., In the Eye of the Beholder: Tort Litigants’ Evaluations of Their Experiences in the Civil Justice System, 24 Law & Soc’y Rev. 953 (1990) (conducting telephone interviews of 122 people whose cases were tried in Fairfax County, Virginia, 74 people whose cases were arbitrated in Bucks County, Pennsylvania, and 90 people in Prince Georges County, Maryland who had participated in judicial settlement conferences); Debra L. Shapiro & Jeanne M. Brett, Comparing Three Processes Underlying Judgments of Procedural Justice: A Field Study of Mediation and Arbitration, 65 J. Personality & Soc. Psychol. 1167 (1993) (conducting telephone interviews with 132 of 158 coal miners, 69 of whom experienced mediation and 89 of whom used arbitration).

  33. [33]. See Shestowsky & Brett, supra note 13, at 66; Stallworth & Stroh, supra note 13. For a description of both studies, see infra Part II.A.3.

  34. [34]. See, e.g., John Thibaut & Laurens Walker, Procedural Justice: A Psychological Analysis 14 (1975); Pauline Houlden et al., Preference for Modes of Dispute Resolution as a Function of Process and Decision Control, 14 J. Experimental Soc. Psychol. 13, 29 (1978) (concluding that the “most preferred procedure corresponds to arbitration”); Stephen LaTour et al., Some Determinants of Preference for Modes of Conflict Resolution, 20 J. Conflict Resol. 319, 349 (finding that arbitration was generally the most preferred procedure). Other research has suggested that adjudicative procedures are regarded as more fair, more satisfying, more accurate, and unbiased from an ex post perspective relative to nonadjudicative ones. E. Allan Lind et al., The RAND Corp., The Perception of Justice: Tort Litigants’ Views of Trial, Court-Annexed Arbitration, and Judicial Settlement Conferences 78–80 (1989).

  35. [35]. See Lind et al., supra note 34, at 74. Thirty-minute telephone interviews were conducted with plaintiffs and defendants in tort cases that had been resolved in the preceding twelve months, in three mid-Atlantic suburban courts. Lind et al., supra note 32, at 954, 961–64. To control for selection bias, each procedure was contrasted with its most common alternative: bilateral settlement. Id. at 960–61. The original analysis, which compared litigants’ evaluations of judicial settlement conferences, trial, and non-binding arbitration, found that ratings of procedural fairness were highest for trial and non-binding arbitration and lowest for settlement conferences. Id. at 961–66. A re-analysis of the data, which compared each third-party procedure to unassisted negotiation, concluded that, compared to negotiation, procedural fairness ratings were higher for trial and arbitration but the same or lower for settlement conferences. Id. at 965–66.

  36. [36]. Larry B. Heuer & Steven Penrod, Procedural Preference as a Function of Conflict Intensity, 51 J. Personality & Soc. Psychol. 700, 707–09 (1986).

  37. [37]. The term “integrative settlement” refers to conflicts that are “variable-sum” (as opposed to “zero sum”) allowing for both parties to “win.” Id. at 706–07.

  38. [38]. Id. at 709; see also Austin et al., supra note 29, at 297 (finding that defendants were least satisfied when an adjudicative procedure yielded an unfavorable outcome, thereby conflicting with previous studies suggesting that adjudicative procedures are the most preferred across all outcome conditions).

  39. [39]. Jeanne M. Brett et al., The Effectiveness of Mediation: An Independent Analysis of Cases Handled by Four Major Service Providers, 12 Negotiation J. 259, 264–65 (1996) (finding higher satisfaction with mediation compared to arbitration, in surveys of disputants and their lawyers). In another study, researchers conducted telephone interviews with coal miners who used either grievance mediation or grievance (non-binding) arbitration. Shapiro & Brett, supra note 32, at 1170­–76. Those who used mediation reported higher satisfaction on measures of procedural justice, control over outcome, and third-party fairness relative to those who used arbitration. Id. at 1175–76.

  40. [40]. Brett et al., supra note 39 (noting respondents favored mediation over arbitration ex post in every aspect tested—“the process and [neutrality], the outcome and its implementation, and the effect of the process on the parties’ relationship”).

  41. [41]. See, e.g., Jennifer E. Shack, Bibliographic Summary of Cost, Pace, and Satisfaction Studies of Court-Related Mediation Programs (2d ed. 2007), available at http://courtadr.org/files/MedStudyBiblio2ndEd2.pdf (reporting on an ambitious examination of studies of a wide variety of ADR programs, primarily in state courts, and concluding that most studies found that mediation participants were more likely to be satisfied with the process and outcome and to find them to be fairer than those who participated in adjudication); David B. Lipsky & Ronald L. Seeber, In Search of Control: The Corporate Embrace of ADR, 1 U. Pa. J. Lab. & Emp. L. 133, 136–41 (1998) (finding that Fortune 1000 employers had a strong preference for mediation over more fact- and law-based procedures such as arbitration).

  42. [42]. Deborah Hensler observed: “The idea that litigants might prefer adversarial processes with the opportunity to adjudicate civil disputes to less adversarial consensual processes . . . is consistent with a long line of social psychological research on individuals’ evaluations of different dispute resolution procedures.” Deborah R. Hensler, Suppose It’s Not True: Challenging Mediation Ideology, 2002 J. Disp. Resol. 81, 81. She continued, “My question is whether legislators’ and judges’ choice of mediation as the procedure that most gratifies these concerns is well grounded. When I look to the findings of the first generation of procedural justice scholars, I see little to support this choice.” Id. at 94. For other examples of this research, see Regina A. Schuller & Patricia A. Hastings, What Do Disputants Want? Preferences for Third Party Resolution Procedures, 28 Canadian J. Behav. Sci. 130, 130–31 (1996), which explained that “a range of experimental studies has consistently found that people prefer adversarial/arbitration type procedures over mediation, since they grant disputants control over the process and involve a binding decision.” See also Tom R. Tyler & E. Allan Lind, Procedural Justice, in Handbook of Justice Research in Law 65, 69–70 (Joseph Sanders & V. Lee Hamilton eds., 2001) (noting that early procedural justice research generally suggested the existence of a preference for adversarial procedures).

  43. [43]. Bingham, supra note 7, at 111 (“In sum, there is a body of field research using a procedural justice framework that finds disputants prefer mediation to the more fact-and law-based alternative of arbitration, at least in the context where both parties have control over dispute system design.”); see also Jean R. Sternlight, Separate and Not Equal: Integrating Civil Procedure and ADR in Legal Academia, 80 Notre Dame L. Rev. 681, 719 (2005) (“Professor Hensler concludes, based on her review of the procedural justice literature that disputants want third-party neutrals to resolve their disputes based on fact and law. Yet, Professor Welsh, after reviewing the same studies, disputes Hensler’s conclusion that disputants view processes as more procedurally fair if they cede decisional control to a third party. Instead, argues Welsh, such studies show that ‘the locus of decision control is less important to litigants’ perceptions of procedural justice than process elements—voice, consideration, even-handedness and dignity.” (footnotes omitted) (quoting Nancy A. Welsh, Disputants’ Decision Control in Court-Connected Mediation: A Hollow Promise Without Procedural Justice, 2002 J. Disp. Resol. 179, 187)).

  44. [44]. See Shestowsky, supra note 6, at 610–16. “Similarly, research evidence suggests that lawyers’ attitudes towards mediation become more positive with time and as a result of repeated experiences with it.” Julie Macfarlane, Will Changing the Process Change the Outcome? The Relationship Between Procedural and Systemic Change, 65 La. L. Rev. 1487, 1498 (2005) (reviewing the relevant research). More experience with ADR procedures is also associated with attorneys being more likely to suggest it to their clients. Roselle L. Wissler, When Does Familiarity Breed Content? A Study of the Role of Different Forms of ADR Education and Experience in Attorneys’ ADR Recommendations, 2 Pepp. Disp. Resol. L.J. 199, 224–26 (2002).

  45. [45]. Shestowsky, supra note 6, at 553 (arguing that disputants might evaluate procedures differently depending on where they are in the dispute resolution trajectory, even for the same case); see also Schuller & Hastings, supra note 42, at 131 (arguing that “preferences and post-satisfaction evaluations may [be affected by] people’s initial perceptions of the consequences that are likely to follow from” each procedure); Shestowsky & Brett, supra note 13, at 63 (reporting research which found that “disputants initially evaluated their options on the basis of the relative control they offered to disputants as opposed to third parties[, but that] . . . initial attraction to disputant control did not predict ex post satisfaction with nonadjudicative procedures”); Tom R. Tyler et al., The Two Psychologies of Conflict Resolution: Differing Antecedents of Pre-experience Choices and Post-experience Evaluations, 2 Group Processes & Intergroup Rel. 99 (1999) (concluding, on the basis of empirical research, that people arrived at ex ante preferences by choosing procedures they believed would help to maximize material outcomes, but based ex post evaluations on the quality of the treatment they received during the procedure, and, in hindsight, were more apt to favor procedures that they felt treated them respectfully and fairly).

  46. [46]. Shestowsky, supra note 6, at 612–16 (“[M]uch of the research supporting a preference for arbitration has assessed ex ante preferences for disputes that would be resolved in the future. In contrast, studies finding a preference for mediation have typically gathered ex post evaluations from disputants who already experienced a procedure to resolve an actual dispute.”).

  47. [47]. Hensler, supra note 42, at 81.

  48. [48]. The procedural justice effect—the fact that citizens care about the process by which outcomes are reached, whether favorable or unfavorable—has generally been found to hold across demographic variables such as race, gender, as well as case variables (such as type of legal issue or amounts in controversy). MacCoun, supra note 12, at 173 (synthesizing the body of literature and noting that the procedural justice effect has been documented across “contexts involving every major demographic category in the United States”); see also Nancy A. Welsh, Making Deals in Court-Connected Mediation: What’s Justice Got to Do with It?, 79 Wash. U. L.Q. 787, 818 (2001) (“[P]erceptions of distributive justice generally have a much more modest impact than perceptions of procedural justice.” (footnote omitted)); id. at 818 n.150 (“[L]aboratory and field studies . . . show that greater perceptions of procedural justice generally produce greater perceptions of distributive justice, regardless of whether the outcome is positive or negative. Occasional studies show that this effect may be reduced when the outcome is positive, but also that this effect continues to be strong when the outcome is negative.” (citations omitted)).

  49. [49]. Donald E. Conlon et al., Nonlinear and Nonmonotonic Effects of Outcome on Procedural and Distributive Fairness Judgments, 19 J. Applied Soc. Psychol. 1085, 1087 (1989) (“One of the most consistent findings in the research on procedural justice is that dispute resolution procedures that provide high process control (i.e., control over presentation of evidence, and the handling of the ‘case’ before a third party) to disputants will enhance perceptions of procedural and distributive fairness.”); Welsh, supra note 48, at 792, 817–22 (synthesizing the research and noting the reliability of the “voice” effect, and the importance of fair treatment and observing that “[p]rocedural justice research indicates clearly that disputants want and need the opportunity to . . . control the telling of [their] story”). But see MacCoun, supra note 12, at 184 (concluding that research shows that fair process matters, but that whether process or outcomes matter more “may not be answerable in a meaningful, global way”).

  50. [50]. John Thibaut et al., Procedural Justice as Fairness, 26 Stan. L. Rev. 1271, 1283 (1974); see also Rebecca Hollander-Blumoff & Tom R. Tyler, Procedural Justice and the Rule of Law: Fostering Legitimacy in Alternative Dispute Resolution, 2011 J. Disp. Resol. 1, 4, 6 (stating that fairness of process is the best predictor of procedural fairness); Russell Korobkin, Psychological Impediments to Mediation Success: Theory and Practice, 21 Ohio St. J. on Disp. Resol. 281, 322 (2006) (“Of particular relevance is the widespread finding that, holding outcomes (especially undesirable ones) constant, people are significantly more satisfied if they rate as ‘fair’ the process that resulted in that outcome.”).

  51. [51]. See supra text accompanying notes 27–33.

  52. [52]. See, e.g., Donna Shestowsky, Disputants’ Preferences for Court-Connected Dispute Resolution Procedures: Why We Should Care and Why We Know So Little, 23 Ohio St. J. on Disp. Resol. 549, 587 (2008) (“Most courts offer a single alternative to trial, typically mediation or arbitration.”); Developments in the LawThe Paths of Civil Litigation, 113 Harv. L. Rev. 1752, 1858 (2000) (“Mediation and arbitration remain the most widespread forms of ADR . . . .”). In federal judicial districts that authorize multiple forms of ADR, sixty-three districts authorize mediation while twenty-three authorize arbitration. Mediation is by far the most common ADR procedure authorized in the federal district courts. Donna Stienstra, Fed. Judicial Ctr., ADR in the Federal District Courts: An Initial Report 6 (2011), available at http://www.fjc.gov/
    public/pdf.nsf/lookup/adr2011.pdf/$file/adr2011.pdf. Some state courts offer arbitration only, or make mediation available but mandate arbitration under particular circumstances. See N.J. Ct. R. 4:21A-1, available at http://www.judiciary.state.nj.us/rules/r4-21a.htm (mandating arbitration for automobile negligence actions and certain other injury actions); Md. Mediation & Conflict Resolution Office, Consumers’ Guide: Alternative Dispute Resolution (ADR) Services in Maryland 77–78 (2013), available at http://www.courts.state.md.us/macro/
    pdfs/consumersguide/consumersguidetoadrservices.pdf (outlining Maryland’s mediation program and noting that “[c]ourts can only order arbitration at the request of all parties”); Cook County Mandatory Arbitration, Circuit Ct. Cook County, http://www.cookcountycourt.org/
    ABOUTTHECOURT/OfficeoftheChiefJudge/CourtRelatedServices/MandatoryArbitration.aspx (follow “Eligibility” hyperlink) (last visited Nov. 17, 2013) (explaining that arbitration is mandated when personal injury, property damage, and breach of contract actions have been brought for $30,000 or less); Court Alternatives, St. N.M. Second Jud. District Ct., http://seconddistrictcourt.nmcourts.gov/calt2.html (last visited Nov. 17, 2013) (mandating arbitration for claims less than or equal to $25,000). Other court systems offer, encourage, or mandate the use of mediation only. See Ky. Rev. Stat. Ann. § 454.011 (West 2006) (“[C]ourts and state governmental agencies are authorized and encouraged to refer disputing parties to mediation before trial or hearing.”); State of N.H. Circuit Court Dist. Div., Civil Writ Mediation Program (2012), available at http://www.courts.state.nh.us/adrp/district/civilrit/
    files/Civil-Writ-Mediationbrochure.pdf (describing the state’s mediation program and explaining that some courts allow litigants to mediate upon request, whereas others send an opt-out notice for litigants who do not want to mediate).

  53. [53]. Stallworth & Stroh, supra note 13, at 33. All disputes involved alleged discrimination of some kind (e.g., age, race, sex, marital status), or disability or sexual harassment issues. Id. at 33–35.

  54. [54]. Id. at 36 tbl.2 (reporting the results of a statistical analysis that found interest in mediation to be significantly stronger than interest in arbitration).

  55. [55]. Id. at 33–34 (“It is not unusual for the Department to take 22 months to ‘commence’ an investigation and three, four, or more years to complete an investigation.”).

  56. [56]. Id. at 34.

  57. [57]. Shestowsky & Brett, supra note 13, at 79–80.

  58. [58]. Id. at 82.

  59. [59]. Participants were drawn from the Cook County Circuit Court in Chicago. Id. at 80-81.

  60. [60]. See id. at 84.

  61. [61]. For a review of the relevant research, see Rubin, supra note 28.

  62. [62]. See Thibaut & Walker, supra note 34.

  63. [63]. Shestowsky & Brett, supra note 13, at 95.

  64. [64]. Id. at 96.

  65. [65]. Id.

  66. [66]. See Heuer & Penrod, supra note 36, at 704; Schuller & Hastings, supra note 42, at 137.

  67. [67]. See Josh A. Arnold & Peter J. Carnevale, Preferences for Dispute Resolution Procedures as a Function of Intentionality, Consequences, Expected Future Interaction, and Power, 27 J. Applied Soc. Psychol. 371, 375–76, 381 (1997).

  68. [68]. See Shestowsky, supra note 16, at 247 (finding that the strength of preference for dispute resolution feature options varied with the participant’s role in the dispute).

  69. [69]. Shestowsky, supra note 6, at 610–11 (discussing past research suggesting that the type of issues (i.e., legal or interpersonal) involved in the dispute influences perceptions of procedures); cf. Neil Vidmar, Procedural Justice and Alternative Dispute Resolution, in Procedural Justice 121, 126 (Klaus F. Rӧhl & Stefan Machura eds., 1997) (summarizing past research showing that the nature of the conflict—for example, claims based on equity considerations versus claims based on legal rights—has an effect on procedural preference and choice).

  70. [70]. See Tom R. Tyler & Heather J. Smith, Social Justice and Social Movements, in 2 The Handbook of Social Psychology 595, 618–19 (Daniel T. Gilbert et al. eds., 4th ed. 1998) (suggesting that although the abstract pursuit of procedural justice is a relatively homogenous goal no matter one’s cultural background, the factors that influence and drive procedural choice—i.e., fairness, respectfulness, just behavior, etc.—may vary across different cultures, races, and ethnicities).

  71. [71]. See, e.g., Kwok Leung & E. Allan Lind, Procedural Justice and Culture: Effects of Culture, Gender, and Investigator Status on Procedural Preferences, 50 J. Personality & Soc. Psychol. 1134, 1138 (1986) (finding that while gender differences did not seem to alter the overall pattern of procedural preferences, there were significant differences in preferences for social influence, persuasion, and mediation, and a marginally significant difference for negotiation).

  72. [72]. The Third Judicial District is composed of Salt Lake County, Summit County, and Tooele County. Together, these counties account for over forty percent of the state’s population (2010 total state population = 2,763,885; 2010 total Third District population = 1,124,197). See State & County QuickFacts, U.S. Census Bureau, http://quickfacts.census.gov/
    qfd/states/49000.html (last modified June 27, 2013) (presenting population for each county separately).

  73. [73]. See Utah Dist. Courts, FY2010 Case Type by Court (2010), available at http://www.
    utcourts.gov/stats/files/2010FY/district/3-Summary.pdf.

  74. [74]. State & County QuickFacts, supra note 72 (select state of California; county of Solano).

  75. [75]. Judicial Council of Cal., 2012 Court Statistics Report: State Caseload Trends: 2001–2002 Through 2010–2011, at 92 tbl.4a (2012).

  76. [76]. Visitors, Multnomah County, http://www.multco.us/visitors (last visited Nov. 17, 2013).

  77. [77]. State & County QuickFacts, supra note 72 (select state of Oregon; county of Multnomah).

  78. [78]. Or. Judicial Dep’t, Statistical Report Relating to the Circuit Courts of the State of Oregon tbl.1 (2010), available at http://courts.oregon.gov/OJD/docs/OSCA/2010_
    Stats_Table_1.pdf.

  79. [79]. A litigant could be both a plaintiff and defendant if the case involves counterclaims.

  80. [80]. The questions were in three different orders to test for order effects. A regression model was generated to test for effects of survey version on litigants’ attraction to procedures. Specifically, attractiveness ratings were regressed on dummy-coded survey version variables to determine whether ratings differed between survey versions. This model was applied to the attractiveness ratings of all eight procedures, and was found to not significantly explain these ratings. The results of these analyses suggest that survey version did not influence attractiveness ratings.

  81. [81]. Litigants whose disputes were designated as foreclosures, family law, landlord-tenant, or bankruptcy were not eligible for the study. See infra Part V.F (discussing case types excluded from the study).

  82. [82]. The process for compiling the case pulls varied slightly by court. For the Utah Court, we logged into the state court’s case filing system and searched for all eligible cases. The Oregon and California Courts sent us a list with litigant names and case names (e.g., “Smith v. Jones”) for all eligible cases.

  83. [83]. The Utah Court sometimes provided litigants’ contact information. In such instances, we used these addresses for the mailing lists. When a pro per litigant’s address was provided by any court, we used that address.

  84. [84]. The mailing lists that were compiled included the litigant’s name, role in the dispute (plaintiff vs. defendant), case name (e.g., “Smith v. Jones”), case file number, filing date, phone number, mailing address, internet databases on which we relied to obtain the contact information, and the research assistant’s “level of certainty” rating for the mailing address.

  85. [85]. When the lead plaintiff or defendant failed to yield a “level of certainty” rating of at least 3, we mailed the survey to the next-listed plaintiff or defendant (using the litigant ordering provided by the court, which was typically based on the complaint) who yielded a rating of at least 3. In some instances, a litigant’s address was narrowed down to two or three possible options with similarly high certainty ratings. For these, letters were sent to each address with the expectation that one of them would reach the correct litigant.

  86. [86]. We did not want to risk litigants on the same side of a case discussing the survey and affecting each other’s responses. Thus, when more than one plaintiff for the same case was randomly drawn to be included in our mailing list, we prioritized the recruitment of only the lead plaintiff (i.e., the person after whom the case was named) or the highest listed litigant if the lead plaintiff was not included in the case pull. The same procedure was followed if multiple defendants for the same case were randomly selected. When a litigant was named in multiple cases filed at the same court in a given case pull, that litigant was mailed a survey for the first numerically listed case number only.

  87. [87]. Initially, the invitation offered participants $25. Starting with the third case pull, the research team increased the offer to $50 and included an opportunity to win a cash prize—one drawing for $500 and one for $750.

  88. [88]. Not all surveys that were returned were included in the dataset. In total, 474 surveys were returned. Several surveys (n = 18) were excluded because the individuals completed more than one survey for the study (for different cases), or their spouse had also completed a survey for the same case (n = 1). Including these surveys would have introduced dependencies in the data. Other surveys were excluded because the survey responses, or other communications from the litigants, revealed that the litigants or their cases ultimately did not match the eligibility requirements (n = 42) (e.g., they completed the survey for a case other than the one for which we solicited them). After these surveys were removed, data from 413 unique litigants remained.

  89. [89]. HLM helps account for dependencies that result from person-specific characteristics when employing repeated-measures designs (e.g., having multiple ratings nested within raters). Stephen W. Raudenbush & Anthony S. Bryk, Hierarchical Linear Models: Applications and Data Analysis Methods 3–4 (2d ed. 2002). By accounting for these dependencies, analyses are derived from more accurate standard errors (“SEs”), which helps to reduce the likelihood of error in the results.

  90. [90]. See, e.g., Jury Instructions Comm. of the Ninth Circuit, A Manual on Jury Trial Procedures 3 (3d ed. 2004), available at http://www.akd.uscourts.gov/docs/general/
    jury_manual.pdf (noting that a civil litigant must take affirmative action to demand a jury trial, the failure to do so constitutes a waiver of a jury trial). For information on the state court rules for the jurisdictions sampled in this study, see Cal. Civ. Proc. Code § 631(f)(4) (West 2013) (“A party waives trial by jury . . . [b]y failing to announce that a jury is required, at the time the cause is first set for trial . . . or within five days after notice of setting . . . .”); Or. Rev. Stat. §52.570 (2013) (“[I]f either party then demands a jury trial and deposits with the justice such trial fee as is required . . . the issue must be tried by a jury and not the justice; but otherwise it must be tried by the justice.”); Utah R. Civ. P. 38(b) (“Any party may demand a trial by jury . . . not later than 10 days after the service of the last pleading directed to such issue.”).

  91. [91]. t (406) = - 2.05, p < .05. A paired t-test was used to compare the ratings for these two procedures. The t-test is a very common statistical test used to compare the means of two groups to determine whether they significantly differ from one another. Andy Field, Discovering Statistics Using SPSS 324 (3d ed. 2009). A paired t-test is used to compare mean ratings made by the same sample. Id. at 325.

  92. [92]. t (409) = - 4.20, p < .001.

  93. [93]. The Chi-square test is a very common and “extremely elegant statistic based on the simple idea of comparing the frequencies you observe in certain categories to the frequencies you might expect to get in those categories by chance.” Field, supra note 91, at 688.

  94. [94]. Attorneys Negotiate with Clients Present tdiff (371) = 6.04, p < .001; Mediation tdiff(365) = 9.00, p < .001; Judge Decides without Trial tdiff (361) = 2.40, p < .05; Jury Trial tdiff(361) = 7.22, p < .001; Judge Trial tdiff (364) = 5.44, p < .001; Binding Arbitration t (363) = 7.82, p<.001; Non-binding Arbitration t (359) = 7.56, p < .001. Attorneys Negotiate without Clients tdiff (375) = .09, ns.

  95. [95]. Multiple regression is a statistical analysis commonly used to predict an outcome (in this case, attractiveness rating for a procedure) based on multiple predictor variables. Field, supra note 91, at 209–10, 790. For each procedure, the intercept of the regression model represents the average attractiveness of the reference group. Thus, significant nominal predictors in the regression model indicate groups within the variable that are associated with a significant change in attractiveness ratings for a procedure compared to the reference group’s average attractiveness rating for that procedure. Similarly, significant continuous predictor variables are variables where changes in the outcome variable correspond significantly with changes in the predictor. The reference group used in the model consisted of individual, White males, between eighteen and twenty-five years of age, who have an individual opposing party with whom they did not have a relationship before their case was filed and with whom they have no interest in having a future relationship, who have a personal injury case, where an insurance company has no interest in the outcome of the case, who have not had experience as either a plaintiff or defendant before, who have zero expectancy of winning at trial, who filed in Oregon, who are plaintiffs in the current case, and who have an extremely negative perception of the court where their case is filed.

  96. [96]. Several steps were performed to keep the number of predictors reasonable: (1) data were collapsed across ethnicities to compare the effect of Whites and Non-Whites; (2) a new variable was created to indicate whether the litigant had been a litigant in a prior case (rather than if he or she had been involved specifically as a plaintiff or defendant); (3) data were collapsed across all options regarding the involvement of insurance companies to compare the effect of either party’s insurance having such an interest and neither party’s insurance having such an interest; and (4) the “other” category for the litigant’s role in the case was excluded from analysis. All nominal variables (e.g., case type) were dummy-coded (see Table H) and continuous variables (e.g., ratings of litigant attraction to Binding Arbitration) were evaluated for assumptions of normality. For those variables found to be high or out of bounds of acceptable ranges used as indicators of normality (see Appendix E), data transformations were applied to bring the variable back in bounds. All regression analyses were conducted by applying the same model to both the transformed and original data for all outcome variables (i.e., ratings of each procedure’s attractiveness). Results from these models were then compared against each other for: (1) concordance in omnibus model significance; and (2) significant differences between the standardized coefficients from the significant omnibus models applied to original versus transformed data. Omnibus model significance conformed very well between the original and transformed data. Furthermore, all standardized coefficients of model predictors generated from raw data were within the 68% CIs of those generated from transformed data, which suggests that the use of transformed data did not result in significantly different predictor effects. Because the unstandardized coefficients generated from the original data are easier to interpret with respect to the original data metrics, the reported results are those generated by applying the regression model to the original data.

  97. [97]. The model explained a significant percentage of the variance in attractiveness ratings for this procedure F(22, 228) = 2.47, p < .001, R = .19.

  98. [98]. The model explained a significant percentage of the variance in attractiveness ratings for this procedure F(22, 226) = 1.87, p < .05, R = .15.

  99. [99]. The model explained a significant percentage of the variance in attractiveness ratings for this procedure F(22, 227) = 1.74, p < .05, R = .15.

  100. [100]. The model explained a significant percentage of the variance in attractiveness ratings for this procedure F(22, 227) = 1.81, p < .05, R = .15.

  101. [101]. The model explained a significant percentage of the variance in attractiveness ratings for this procedure F(22, 226) = 1.75, p < .05, R = .15.

  102. [102]. The model explained a significant percentage of the variance in attractiveness ratings for this procedure F(22, 227) = 2.46, p < .001, R = .19.

  103. [103]. In this Article, a “repeat-litigant” is one who acted as either a plaintiff or defendant in at least one case prior to the one for which they completed the survey for this study.

  104. [104]. F(22, 226) = 1.30, p = .18, R = .11.

  105. [105]. F(22, 227) = 1.04, p = .42, R= .09.

  106. [106]. See supra Part IV.

  107. [107]. See supra Part IV.A.

  108. [108]. HLM analysis using Attorneys Negotiate with Clients Present as the reference group confirmed this conclusion. The results of this analysis are on file with the Author.

  109. [109]. See, e.g., Stephen Daniels, The Question of Jury Competence and the Politics of Civil Justice Reform: Symbols, Rhetoric, and Agenda-Building, Law & Contemp. Probs., Autumn 1989, at 269 (calling for tort reform while criticizing the civil justice system); Elizabeth Loftus, Insurance Advertising and Jury Awards, 65 A.B.A. J. 68, 69 (1979) (reporting evidence of an effective, multi-million-dollar media campaign by the insurance industry to influence jury awards nationwide); Richard Waites & Jim Lawrence, Juror Perceptions About Lawsuits and Tort Reform, Advocates,


    http://www.theadvocates.com/Juror%20Perceptions%20About%20Lawsuits%20and%20Tort%20Reform.pdf (last visited Nov. 17, 2013) (stating that the prevalence of negative attitudes towards juries may be due “to the activities of people and groups who support civil justice reform”); Am. Tort Reform Ass’n, http://www.atra.org (last visited Nov. 17, 2013);
  110. [110]. See Kevin M. Clermont & Theodore Eisenberg, Litigation Realities, 88 Cornell L. Rev. 119, 131 (2002) (explaining that while “most commentators have assumed that the wait in the jury queue [is] longer than the wait for a judge’s trial and decision,” research on federal courts suggests that “the reality is the opposite”).

  111. [111]. See supra notes 36–41 and accompanying text.

  112. [112]. The “shuttle” model of mediation was not included in the description of Mediation that was provided to the participants. See infra Appendix D. Shuttle mediation occurs when mediators meet with the parties separately rather than in joint session and “shuttle” information back and forth between the parties in an effort to reach an agreement. See Shuttle Mediation, Scottish Community Mediation Centre, http://www.scmc.sacro.org.uk/SCMC_12_Shuttle
    Mediation.pdf (last visited Nov. 17, 2013).

  113. [113]. H. Warren Knight et al., California Practice Guide: Alternative Dispute Resolution § 2:54 (2012) (“It is usually advantageous for the client not to be present during settlement negotiations.”); David A. Hoffman, Mediation and the Art of Shuttle Diplomacy, 27 Negotiation J. 263, 303 (2011) (“Outside the presence of their clients, the lawyers tend to be more candid, and the conversation can proceed more efficiently because the lawyers do not feel as much need to impress the clients or the opposing party.”); Laura A. Kaster, Improving Lawyer Judgment by Reducing the Impact of “Client-Think, Disp. Resol. J., Feb.–Apr. 2012, at 56, 58 (arguing that clients may negatively affect settlement outcomes). But see Shawn P. Davisson, Note, Privatization and Self-Determination in the Circuits: Utilizing the Private Sector Within the Evolving Framework of Federal Appellate Mediation, 21 Ohio St. J. on Disp. Resol. 953, 987–88 (2006) (discussing how having a client present allows lawyers to posture and therefore reach better results).

  114. [114]. See supra Part IV.A.

  115. [115]. Ari Davis, Moving from Mandatory: Making ADR Voluntary in New York Commercial Division Cases, 8 Cardozo J. Conflict Resol. 283, 295 (2006) (“Court-mandated ADR is generally better attended than voluntary ADR programs.”); Schuller & Hastings, supra note 42, at 130 (noting that arbitration and mediation “have been associated with low voluntary usage”); Alexandria Zylstra, The Road from Voluntary Mediation to Mandatory Good Faith Requirements: A Road Best Left Untraveled, 17 J. Am. Acad. Matrimonial L. 69, 79 (2001) (“[M]andatory statutes are necessary to overcome low usage rates and other pitfalls . . . .”).

  116. [116]. See supra Part IV.A.

  117. [117]. See generally John Lande, The Movement Toward Early Case Handling in Courts and Private Dispute Resolution, 24 Ohio St. J. on Disp. Resol. 81, 99–101 (2008) (arguing that Early Neutral Evaluation is promising for both judicial efficiency and early dispute resolution).

  118. [118]. See supra Part IV.B.

  119. [119]. See supra Part IV.C.

  120. [120]. See supra Part IV.A.

  121. [121]. See, e.g., Welsh, supra note 48, at 817; Nancy A. Welsh, The Current Transitional State of Court-Connected ADR, 95 Marq. L. Rev. 873, 877–84 (2012) (reviewing responses from “lawyers, judges, mediators, administrators, policymakers, funders, and academics” about the “value-added” to courts by ADR and what must happen to make ADR “good”).

  122. [122]. See infra Appendix D.

  123. [123]. See supra Part IV.C.

  124. [124]. See supra Part IV.C.

  125. [125]. Ostrom et al., supra note 3, at 768–70 (reporting on an impressive examination of state court cases across twenty-two states, and concluding that, in 2002, the rate of jury trials as a proportion of civil dispositions was .6%, and the rate of bench trials as a proportion of civil dispositions was 15%); see also Paula Hannaford-Agor et al., Trial Trends and Implications for the Civil Justice System, Caseload Highlights, June 2005, at 1, available at http://contentdm.
    ncsconline.org/cgi-bin/showfile.exe?CISOROOT=/civil&CISOPTR=25 (“[O]ver the past two decades . . . the ratio of bench to jury trials has remained fairly stable––approximately 26 to 1.”).

  126. [126]. Ostrom et al., supra note 3, at 768–70. Similar data were available for two of the study courts. These data also support the view that most filed cases are resolved outside of trial. See Judicial Council of Cal., 2012 Court Statistics Report: Statewide Caseload Trends: 2001–2002 Through 2010–2011, at 75 (2012), available at http://www.courts.ca.gov/
    documents/2012-Court-Statistics-Report.pdf (finding that of 952,309 civil dispositions within the fiscal year 2010–2011, only 8% of those were disposed of after trial, while the rest were disposed of before trial); see also Office of the State Court Adm’r, Supreme Court of Or., Statistical Report Relating to the Circuit Courts of the State of Oregon: Second Half 2011, at tbl.6, available at http://courts.oregon.gov/OJD/OSCA/pages/2011
    statistics.aspx (containing statistics of each county in Oregon in 2011, and showing that in most counties, less than 5% of civil cases went to trial).

  127. [127]. The high level of confidence that litigants expressed in “winning” if their cases went to trial supports this interpretation. See infra text accompanying notes 141–43. Such confidence might lead them to believe that they would fare better at trial than by accepting any settlement offer.

  128. [128]. The research team is following each case to conduct this analysis for a future publication.

  129. [129]. See supra Parts IV.C–D.

  130. [130]. Shestowsky & Brett, supra note 13, at 85–88.

  131. [131]. See supra Part IV (discussing results).

  132. [132]. The research team evaluated answers to this question, the data for which are intended for a future publication. The team developed a coding scheme that reflected the range of responses. Two law-student coders then independently coded each litigant’s response. The Author compared the two sets of coding results and determined the instances in which the coders disagreed. The coders then deliberated to resolve these disagreements. All disagreements were resolved. The reported calculations were made using the resulting dataset.

  133. [133]. For statistical reasons explained earlier, to draw this conclusion, only the procedures for which the regression model was found to be statistically significant were considered. See supra Part IV.E.Mediation and Non-binding Arbitration.

  134. [134]. See supra text accompanying note 102.

  135. [135]. See, e.g., Lisa B. Bingham, Employment Arbitration: The Repeat Player Effect, 1 Emp. Rts. & Emp. Pol’y J. 189 (1997); Lisa B. Bingham, On Repeat Players, Adhesive Contracts, and the Use of Statistics in Judicial Review of Employment Arbitration Awards, 29 McGeorge L. Rev. 223, 232–35 (1998).

  136. [136]. Joseph L. Forstadt, Discovery in Arbitration, in ADR & the Law 52, 52 (Am. Arbitration Ass’n et al. eds., 20th ed. 2006) (“One of the primary ways in which arbitration is less costly, both in terms of time and money, is that it normally has less extensive discovery than traditional litigation.”).

  137. [137]. Jennifer Walker Elrod, Is the Jury Still Out?: A Case for the Continued Viability of the American Jury, 44 Tex. Tech L. Rev. 303, 318 n.122 (2012) (“[N]oting that ‘[c]onventional wisdom suggests that businesses choose binding arbitration mainly because it is perceived to be different from litigation,’ providing benefits such as ‘cost savings, shorter resolution times,’ and ‘expert decision makers.’” (quoting Thomas J. Stipanowich, Arbitration: The “New Litigation,2010 U. Ill. L. Rev. 1, 4)).

  138. [138]. See Coll. of Commercial Arbitrators, Protocols for Expeditious, Cost-Effective Commercial Arbitration 6 (Thomas J. Stipanowich et al. eds., 2010) (“Although many arbitrators and some arbitration rules aim to hold the line on excessive discovery, it is not unusual for legal advocates to agree to litigation-like procedures for discovery, even to the extent of employing standard civil procedural rules.”); Forstadt, supra note 136, at 56–57.

  139. [139]. See Thomas J. Stipanowich, The Arbitration Fairness Index: Using a Public Rating System to Skirt the Legal Logjam and Promote Fairer and More Effective Arbitration of Employment and Consumer Disputes, 60 U. Kan. L. Rev. 985, 988–91 (2012) (recognizing the myriad reasons for which mandatory arbitration in employment and consumer disputes operates unfairly toward individual disputants).

  140. [140]. A t-test compared expected use ratings for Binding Arbitration for litigants opposing individuals (n = 179, M = 3.15, SE = .19) and those opposing companies (n = 140, M = 3.90, SE = .23), t (317) = -2.15, p < .05. Litigants who indicated that they were opposing both an individual and company (n= 3) were excluded from the analysis.

  141. [141]. See supra text accompanying notes 99–101.

  142. [142]. The study courts use various forms of non-binding arbitration, which produce outcomes that are adjudicated, but that either party can veto. For instance, in Oregon, arbitration can be either mandatory or voluntary. What is Arbitration?, Or. Jud. Dep’t, http://courts.oregon.gov/OJD/programs/adr/pages/whatisarbitration.aspx (last visited Nov. 17, 2013). In the latter, “the parties must agree in advance . . . [on] how parties can appeal an arbitrator’s decision.” Id. Under Oregon’s voluntary arbitration scheme, parties may appeal pursuant to Or. Rev. Stat. § 36.425(2) (2013). In Utah, the court refers general civil matters to an ADR program. Mediation seems to be the court’s preferred procedure, but the parties can stipulate to “binding” or non-binding arbitration. See Utah Code Ann. § 78B-6-206 (West 2013); Utah R. Ct.-Annexed Alt. Disp. Resol. R. 102. Although Utah’s ADR program contemplates binding arbitration in the case of motor vehicle injury, the code allows litigants to request a trial de novo after the final arbitration judgment is issued, thereby rendering the arbitration functionally non-binding. See generally Utah Code Ann. § 31A-22-321(11) (West 2013). Finally, in California, rules pertaining to arbitration are promulgated at the local level. For instance, in the California Court from which litigants were selected for the present study, Rule 4.7 governs diversion to arbitration. Under this rule, civil actions in which the amount in controversy does not exceed $50,000 “shall be subject to judicial nonbinding arbitration.” Solano Cnty. Ct. R. 4.7(a)(1)(a).

  143. [143]. This interpretation resonates with research suggesting that people arrive at pre-experience preferences for procedures by choosing ones that help them to maximize self-interest in terms of material outcomes. See Tyler et al., supra note 45. It is also consistent with lab research suggesting that when participants have a strong case, they favor procedures in which a third party has decision control. Heuer & Penrod, supra note 36, at 704 (reporting on laboratory research in which they found “unequivocal support” for the notion “that disputants with a strong case . . . prefer the autocratic and arbitration procedures, whereas their weak-case counterparts . . . prefer the moot, mediation, and bargaining procedures”); Schuller & Hastings, supra note 42, at 135–36 (participants were assigned a role that was either “strong” or “weak” on the facts; those with the strong case indicated that they would choose adjudication or binding arbitration over mediation; whereas those with the weak case said they would select binding arbitration over adjudication, but rated mediation in between these two options). But see Thibaut et al., supra note 50, at 1280–82 (describing laboratory research in which the roles that participants played were varied according to relative advantage based on the facts of the case and finding that although the adversary model was preferred by all participants, those advantaged by the facts liked the inquisitorial and single investigator procedures more than would those in the disadvantaged role, and individuals who were disadvantaged by the facts liked the adversary procedure more than would those who were advantaged by them, possibly “because the adversary procedure could be expected to enable the disadvantaged party to bolster his case and dispute his opponent’s contentions through the services of a competent legal representative sympathetic to his position”).

  144. [144]. In a representative study demonstrating the egocentric bias in litigation contexts, participants were assigned to the role of plaintiff or defendant and asked to estimate what a judge would award based on the same information. They then negotiated the resolution of the case. Regardless of their role in the case, participants believed that a judge would more likely rule in their favor, suggesting that, on average, judgments of how much they would get from a judge if negotiations failed were inflated relative to what objective assessments would suggest. George Loewenstein et al., Self-Serving Assessments of Fairness and Pretrial Bargaining, 22 J. Legal Stud. 135, 149–53 (1993).

  145. [145]. n = 398.

  146. [146]. See supra text accompanying note 99.

  147. [147]. Some studies have found that women also exhibit enhanced concern for the other party, more willingness to make concessions, and a preference for collaborative strategies. See, e.g., Kwok Leung et al., Effects of Cultural Femininity on Preference for Methods of Conflict Processing: A Cross-Cultural Study, 26 J. Experimental Soc. Psychol. 373, 384–87 (1990); Christine Rack, Negotiated Justice: Gender & Ethnic Minority Bargaining Patterns in the MetroCourt Study, 20 Hamline J. Pub. L. & Pol’y 211, 220–24 (1999).

  148. [148]. Repeated Measures ANOVA was used to test whether gender (Male n = 225; Female n = 172) was related to changes in attractive ratings for the Jury Trial, the Judge Trial, and Binding Arbitration. Mauchly’s test of sphericity was significant (χ(2) = 26.97, p < .01), so the Greenhouse-Geisser (ε = .94) correction was used. The interaction between gender and procedures rated was significant, F(1.86, 740.98) = 4.32, p = .02. The interaction was further evaluated using pair-wise comparisons adjusted using the Bonferroni correction.

  149. [149]. Binding Arbitration: MDiff = 1.28, SE = .24, p < .01; Jury Trial: MDiff = .62, SE = .26, p = .02; Judge Trial: MDiff = .40, SE = .23, ns.

  150. [150]. Judge Trial (M = 5.92, SE = .15), Jury Trial (M = 5.45, SE = .17), MDiff = .47, SE = .19, p = .04; Judge Trial (M = 5.92, SE = .15), Binding Arbitration (M = 4.93, SE = .16), MDiff = .99, SE = .19, p < .01.

  151. [151]. Judge Trial (M = 5.52, SE = .17), Jury Trial (M = 4.83, SE = .20), MDiff = .69, SE = .22, p < .01. Judge Trial (M = 5.52, SE = .17), Binding Arbitration (M = 3.66, SE = .18), MDiff = 1.86, SE = .22, p < .01.

  152. [152]. Jury Trial (M = 5.45, SE = .16), Binding Arbitration (M = 4.93, SE = .16), MDiff = .52, SE = .23, p = .07.

  153. [153]. Jury Trial (M = 4.83, SE = .20), Binding Arbitration (M = 3.66, SE = .18), MDiff = 1.17, SE = .26, p < .01.

  154. [154]. See supra text accompanying note 99.

  155. [155]. See Alissa J. Strong, “But He Told Me It Was Safe!”: The Expanding Tort of Negligent Misrepresentation, 40 U. Mem. L. Rev. 105, 142 (2009) (“A problem that arises in personal injury cases is that juries sympathize with and strongly desire to compensate the victim.”). But see, e.g., Philip G. Peters, Jr., Hindsight Bias and Tort Liability: Avoiding Premature Conclusions, 31 Ariz. St. L.J. 1277, 1293 (1999) (“It is widely believed that plaintiffs benefit from jury sympathies. Yet, an increasing body of evidence suggests that jurors begin their job favoring tort defendants and doubting the motives of personal injury plaintiffs . . . .”).

  156. [156]. Specifically, an interaction term was added hierarchically to the regression model to test whether being a plaintiff versus defendant moderated the relation between litigant attraction to jury trials and confidence in winning their personal injury case. This analysis tested whether the confidence ratings of litigants involved in personal injury cases had a consistent relation with attraction to the Jury Trial for both plaintiffs and defendants. Although it is worth noting that the available sample size for the test was relatively small, there was no significant change in the variance of attraction to jury trials explained by the addition of the interaction term, ∆R = .04, F(1,50) = 3.21, Fcrit = 4.03, p = .08, This result suggests that, for litigants involved in personal injury cases, being a plaintiff versus a defendant did not moderate the relation between attraction to jury cases and confidence in winning the case.

  157. [157]. See supra text accompanying notes 97–98.

  158. [158]. Some empirical research suggests that mediation has a less negative impact on underlying relationships between the parties than trial. See Craig A. McEwen & Richard J. Maiman, Small Claims Mediation in Maine: An Empirical Assessment, 33 Me. L. Rev. 237, 257 (1981); Roselle L. Wissler, Mediation and Adjudication in the Small Claims Court: The Effects of Process and Case Characteristics, 29 Law & Soc’y Rev. 323, 347–48 (1995). But see Richard J. Maiman, An Evaluation of Selected Mediation Programs in the Massachusetts Trial Court 7–9, 35, 37 (1997) (finding that litigants were as likely to think that mediation had not improved their relationship as to think that it had).

  159. [159]. See, e.g., Tyler, supra note 10, at 57–68 (presenting original data showing that legitimacy has more influence on obedience than deterrence or peer disapproval); Tom R. Tyler & Kenneth Rasinski, Procedural Justice, Institutional Legitimacy, and the Acceptance of Unpopular U.S. Supreme Court Decisions: A Reply to Gibson, 25 Law & Soc’y Rev. 621 (1991). But see Jeffery J. Mondak, Institutional Legitimacy and Procedural Justice: Reexamining the Question of Causality, 27 Law & Soc’y Rev. 599, 604–07 (1993) (reporting on an experimental study that failed to show that causality runs from procedural justice to institutional legitimacy).

  160. [160]. See supra Part V.D.3.

  161. [161]. Due to the small sample size of Non-White groups, Non-White individuals were collapsed into a single Non-White group to have acceptable group sizes for comparisons. Thus, distinct Non-White sub-group (e.g., African American or Asian) comparisons against White or other groups cannot be discerned.

  162. [162]. A curvilinear effect of age was not examined.

  163. [163]. See supra Part II.A.

  164. [164]. See supra Part II.A.

  165. [165]. See e.g., Steven J. Spencer et al., Establishing a Causal Chain: Why Experiments Are Often More Effective than Mediational Analyses in Examining Psychological Processes, 89 J. Personality & Soc. Psychol. 845, 850–51 (2005) (emphasizing “the power of experiments” for “demonstrating causality”); see also Dennis Garlick, Understanding the Nature of the General Factor of Intelligence: The Role of Individual Differences in Neural Plasticity as an Explanatory Mechanism, 109 Psychol. Rev. 116, 125–26 (2002) (arguing that independently manipulating the possible causal factor is necessary to establish possible causation); Elizabeth P. Shulman et al., Moral Disengagement Among Serious Juvenile Offenders: A Longitudinal Study of the Relations Between Morally Disengaged Attitudes and Offending, 47 Developmental Psychol. 1619, 1621, 1629–31 (2011) (explaining why nonexperimental, correlational studies “cannot provide definitive evidence of causality”).

  166. [166]. See, e.g., James S. Kakalik et al., An Evaluation of Mediation and Early Neutral Evaluation Under the Civil Justice Reform Act 24–25 (1996), available at http://
    www.rand.org/content/dam/rand/pubs/monograph_reports/2007/MR803.pdf (“Complete responses to our surveys were received from . . . about one-ninth of the litigants on closed cases (about one-fifth of the litigants on closed cases for whom we had addresses).”); James S. Kakalik et al., Discovery Management: Further Analysis of the Civil Justice Reform Act Evaluation Data, 39 B.C. L. Rev. 613, 618 (1998) (noting a 13% response rate for mail study of litigants’ perceptions of court proceedings); William H. Schwab, Collaborative Lawyering: A Closer Look at an Emerging Practice, 4 Pepp. Disp. Resol. L.J. 351, 371 (2004) (noting that 7.1% of disputants returned a completed mail survey about collaborative lawyering); Shestowsky & Brett, supra note 13, at 80–81 (reporting a 6.4% response rate for surveys mailed to civil litigants with cases pending in the Circuit Court of Cook County, Illinois); Stallworth & Stroh, supra note 13, at 33–34 (noting that questionnaires were mailed to 3000 parties with cases pending at the Illinois Human Rights Commission; 211 parties responded). But see Lamont Stallworth et al., The NLRB’s Unfair Labor Practice Settlement Program: An Empirical Analysis of Participant Satisfaction, Disp. Resol. J., Nov. 2004–Jan. 2005, at 22, 25 (obtaining a 28% response rate for disputants using a mail survey to obtain perceptions of a settlement program).

  167. [167]. See Shestowsky & Brett, supra note 13, at 80–81 (reporting a 6.4% response rate for surveys mailed to civil litigants with cases pending in the Circuit Court of Cook County, Illinois); Stallworth & Stroh, supra note 13, at 33–34 (reporting that questionnaires were mailed to 3000 parties with cases pending at the Illinois Human Rights Commission; 211 parties responded).

  168. [168]. See Christopher Jepson et al., In a Mailed Physician Survey, Questionnaire Length Had a Threshold Effect on Response Rate, 58 J. Clinical Epidemiology 103, 104–05 (2005) (noting that surveys under 1000 words were returned at a rate of 59.4% whereas surveys over 1000 words were returned at a rate of 38.0%). But see Jagdish N. Sheth & A. Marvin Roscoe, Jr., Impact of Questionnaire Length, Follow-Up Methods, and Geographical Location on Response Rate to a Mail Survey, 60 J. Applied Psychol. 252, 252–53 (1975) (comparing response rates between a four-page survey designed to take ten minutes to complete and a six-page survey designed to take eighteen minutes to complete, and finding no difference in response rate).

  169. [169]. The original goal of the study was to include at least one court from both the East and West Coast. An eight-month-long search for an East Coast court that met the study criteria that would also provide remote access to case disposition information (which was needed for another part of the project) was not successful.

  170. [170]. (1) = 15.96, p < .01. Individuals who indicated that they were both a plaintiff and a defendant (n = 12) or neither (n = 1) were excluded from the analysis.

  171. [171]. A binomial test was used to determine if the proportion of participants who were plaintiffs (56.9% of 413) significantly differed from the proportion of all litigants sent recruitment letters who were plaintiffs (44.1%). The test revealed that there was a significant difference (p < .05) between the two proportions, suggesting that the ratio of plaintiffs to defendants was significantly larger in the actual sample compared to the ratio in the sample used for recruitment.

  172. [172]. See, e.g., 20 Am. Jur. 2d Courts § 13 (2012) (defining the term “special court”); 17 Charles Allen Wright et al., Federal Practice & Procedure § 4102 (3d ed. 2012) (outlining the jurisdiction of the federal tax courts); id. § 3570 (outlining the jurisdiction of the federal bankruptcy courts); Tom R. Tyler, Multiculturalism and the Willingness of Citizens to Defer to Law and to Legal Authorities, 25 Law & Soc. Inquiry 983, 984 (2000) (noting that landlord­–tenant disputes are handled in small-claims courts).

  173. [173]. Tom R. Tyler, Citizen Discontent with Legal Procedures: A Social Science Perspective on Civil Procedure Reform, 45 Am. J. Comp. L. 871, 876 (1997).

  174. [174]. Utah Code Ann. § tit. 78B, ch. 6, pt. 2 (West 2013). The Alternative Dispute Resolution Act mandates the Judicial Council to implement a program utilizing Alternative Dispute Resolution measures in state courts. Id. § 78B-6-203. The program was implemented by the Judicial Council and Utah Supreme Court rules on January 1, 1995. James R. Holbrook & Laura M. Gray, Court-Annexed Alternative Dispute Resolution, 21 J. Contemp. L. 1, 15-16 (1995); see also Utah Code Ann. § 78B-6-205.

  175. [175]. Utah Code Jud. Admin. R. 4-510 (2011) (amended 2012). This Rule applies to all civil cases filed in Utah’s Second, Third, and Fourth Judicial Districts. Id.

  176. [176]. The cost of ADR procedures is typically borne by the disputants, but disputants may contact the ADR Director to request a pro bono mediator if they are impecunious. Id. at 4-510(12).

  177. [177]. ADR procedures do not apply to:


    (1) Title 26, Chapter 19, Medical Benefits Recovery Act;
    (2) Title 62A, Chapter 11, Recovery Services;
    (3) Title 78A, Chapter 8, Small Claims Court;
    (4) Title 78B, Chapter 6, Part 8, Forcible Entry and Detainer;
    (5) Title 78B, Chapter 7, Part 1, Cohabitant Abuse Act;
    (6) Title 78B, Chapter 12, Utah Child Support Act;
    (7) Title 78B, Chapter 15, Utah Uniform Parentage Act;
    (8) Title 78B, Chapter 13, Utah Uniform Child Custody Jurisdiction and Enforcement Act;
    (9) Title 62A, Chapter 15, Substance Abuse and Mental Health Act;
    (10) Rules 65A, 65B and 65C of the Utah Rules of Civil Procedure;
    (11) temporary orders requested under Title 30, Husband and Wife;
    (12) uncontested matters brought under:
    (12)(A) Title 42, Chapter 1, Change of Name;
    (12)(B) Title 75, Utah Uniform Probate Code;
    (12)(C) Title 78B, Chapter 5, Part 3, Foreign Judgment Act;
    (12)(D) Title 78B, Chapter 6, Part 1, Adoption; or
    (13) actions pursued by an assignee of a claim.

    Id. at 4-510.

  178. [178]. See id. at 4-510(6).

  179. [179]. Id. at 4-510(6)(A).

  180. [180]. “All parties file with the clerk a written agreement signed by counsel and the parties to submit the case to nonbinding arbitration pursuant to URCADR Rule 102.” Id. at 4-510(6)(A)(ii).

  181. [181]. “All the parties file with the clerk a written agreement signed by counsel and the parties to submit the case to binding arbitration as provided by law.” Id. at 4-510(6)(A)(iii) (2011) (amended 2012).

  182. [182]. Utah R. Ct.-Annexed Alt. Dis. Resol. R. 101(a), 102(a); Utah Code Jud. Admin. R. 4-510(11); see also Utah Courts Mediation Program, Utah St. Courts, http://www.utcourts.gov/
    mediation/roster/index.asp (last modified Mar. 11, 2013).

  183. [183]. “At any time . . . upon its own motion, or for good cause shown upon motion by a party, the court may order that an action that has been referred to the ADR program be withdrawn from the ADR program and restored to the trial calendar.” Utah Code Jud. Admin. R. 4-510(7)(B).

  184. [184]. See Utah R. Civ. P. 41. This “traditional right to dismiss” means that the case is no longer stayed. No separate carve-out for dispositional motions occurs.

  185. [185]. Cal. Bus. & Prof. Code § 465(a) (West 2013). The Dispute Resolution Programs Act of 1986 is premised on the Legislature’s finding that “[t]he resolution of many disputes can be unnecessarily costly, time-consuming, and complex when achieved through formal court proceedings where the parties are adversaries and are subjected to formalized procedures.” See id.§ 465(a). The Act provides for the local establishment and funding of informal alternative dispute resolution programs, but does not mandate specific ADR requirements. Rather, the Act encourages local courts to create their own ADR procedures. See id. § 465(c)-(e).

  186. [186]. In 2011, the California Court’s rules provided that at a case’s first Case Management Conference all matters shall be diverted to arbitration where: “(1) The parties stipulate to arbitration; (2) The plaintiff requests arbitration; (3) The amount in controversy does not exceed $50,000 as to any plaintiff; or, (4) Arbitration might reasonably lead to the resolution of the case.” Ca. R. Solano Super. Ct. 4.7 (2011) (amended 2012), available at http://www.
    solano.courts.ca.gov/materials/Local%20Rules%20--%20eff%2007-01-11%20-%20Complete.
    pdf. However, communications with the court revealed that since the court does “not offer a free ADR Program, the court will not generally order parties to participate in a program that they must pay for . . . ADR is not mandated or required. Parties may simply choose not to participate in ADR.” E-mail from Grace A. Andres, ADR Adm’r & Court Serv. Program Manager, Civil & Small Claims Divs., Superior Court of Cal., Cnty. of Solano, to Shannon Clawson, Research Assistant to author (Dec. 1, 2009, 16:39 PST) (on file with author). However, judges have discretion, on a case-by-case basis, to order “no fee mediation” if they deem it appropriate. To support these instances, mediators on the court’s roster sign an agreement to provide eight hours of free mediation. In addition, it is most likely that “Rule 4.14 (e) [sic] was written to cover any number of circumstances that may apply to a particular case when the parties do want to participate, but their case may not meet all the requirements of the rule of court. It gives the judge the ability to waive a requirement if the parties do want to participate and it appears there is a chance of settlement through ADR.” Id.

  187. [187]. E-mail from Grace A. Andres to Shannon Clawson, supra note 186.

  188. [188]. Compare Utah Code Jud. Admin. R. 4-510 (2011) (amended 2012), with E-mail from Grace A. Andres to Shannon Clawson, supra note 186.

  189. [189]. E-mail from Grace A. Andres, ADR Adm’r & Court Serv. Program Manager, Civil & Small Claims Divs., Superior Court of Cal., Cnty. of Solano, to Shannon Clawson, Research Assistant to author (Dec. 3, 2009, 14:09 PST) (on file with author).

  190. [190]. In this context “divert” means that “the case is not stayed pending completion of mediation, and a review date or further case management hearing is set.” E-mail from Grace A. Andres, ADR Adm’r & Court Serv. Program Manager, Civil & Small Claims Divs., Superior Court of Cal., Cnty. of Solano, to Saba Shatara, Research Assistant to author (Oct. 15, 2012, 08:21 PST) (on file with author).

  191. [191]. E-mail from Grace A. Andres to Shannon Clawson, supra note 189.

  192. [192]. E-mail from Grace A. Andres, ADR Adm’r & Court Serv. Program Manager, Civil & Small Claims Divs., Superior Court of Cal., Cnty. of Solano, to Shannon Clawson, Research Assistant to author (Apr. 6, 2010, 10:10 PST) (on file with author).

  193. [193]. See Cal. R. Solano Super. Ct. 4.6(a), (f)(2) (2011) (amended 2012). These conferences are set for the week in which the 120th day from the filing of the complaint transpires. Id. at 4.6(f)(1).

  194. [194]. Id. at 4.6(f)(2); E-mail from Grace A. Andres to Shannon Clawson, supra note 189.

  195. [195]. Cal. R. Solano Super. Ct. 4.8(d)(1).

  196. [196]. According to the version of the Solano County Court rules in force at the time of this study, “[u]pon ordering the matter to nonbinding arbitration, the court will direct that the arbitration be concluded within ninety (90) days, unless the court determines in its discretion that a longer timeframe is appropriate.” Id. at 4.7(b). The Local Rules did not specify a time limit for commencing mediation. See id. at 4.8.

  197. [197]. The arbitrator need not be a judge. Pursuant to Oregon’s Uniform Trial Court Rules section 13.090:


    (1) Unless otherwise ordered or stipulated, an arbitrator must be an active member in good standing of the Oregon State Bar, who has been admitted to any Bar for a minimum of five years, or a retired or senior judge. The parties may stipulate to a nonlawyer arbitrator.
    (2) An arbitrator who is not a retired or senior judge or stipulated nonlawyer arbitrator must be an active member in good standing of the Oregon State Bar at the time of each appointment.

    Or. Unif. Trial Ct. R. 13.090 (2012).

  198. [198]. Or. Rev. Stat. § 36.400(3) (2012).

  199. [199]. See id. § 36.405(1)(a).

  200. [200]. Multnomah Cnty. Local R. 12.025 (2013); E-mail from Doug Bray, Trial Court Adm’r, Multnomah Cnty. Circuit Court, to Shannon Clawson, Research Assistant to author (May 09, 2010, 18:49 PST) (on file with author).

  201. [201]. Or. Unif. Trial Ct. R. 5.150(1) (“A civil case eligible for jury trial may be designated as an expedited case.”), implemented by In the Matter of Out-of-Cycle adoption of New UTCR 5.150, UTCR Form 5.150.1a, and UTCR Form 5.150.1b, Chief Justice Order No. 10-025 (Or. May 6, 2010). As of July 2013, only fifteen cases had applied for this option. E-mail from Nan G. Waller, Presiding Judge, Multnomah Cnty. Circuit Court, to Adrienne Nelson, Circuit Judge, Multnomah Cnty. Circuit Court (July 15, 2013, 16:39 PST) (on file with author).

  202. [202]. Or. Unif. Trial Ct. R. 5.150(2)(b).

  203. [203]. Local Rules require “[w]ithin 14 days after notification by the court that the case is assigned to arbitration, any party seeking exemption from arbitration must file and serve a ‘Motion for Exemption from Arbitration.’” Id. at 13.070.

  204. [204]. Or. Rev. Stat. § 36.405(2).

  205. [205]. Summary Statement on Arbitration in the Fourth Judicial District from Doug Bray, Trial Court Adm’r, Multnomah Cnty. Circuit Court [hereinafter Summary Statement from Doug Bray] (on file with author) (noting that the motions for removal from arbitration upon good cause “are routinely granted and the case is restored to the circuit court’s jury trial track for civil actions”). The standard for good cause requires “an actual compelling reason” for removal, more than a mere “on request” standard. E-mail from Doug Bray to Shannon Clawson, supra note 200. The court estimates that:

    approximately 10 percent of filed civil actions in the Fourth Judicial District are initially referred to the mandatory arbitration program. Many of those cases, at least 25 percent, are removed from the mandatory arbitration program either because of amendments to the claims or counterclaims which take the matter out of the jurisdiction of arbitration program, the subject matter of the action is in fact not within the scope of the mandatory referral to arbitration statute, or are removed from arbitration on application of the parties for good cause shown.

    Summary Statement from Doug Bray, supra.

  206. [206]. See Or. Rev. Stat. § 36.405(3) (2011). The Oregon Court requires the arbitration hearing “to take place not later than 91 days . . . from the date of assignment to arbitration.” Multnomah Cnty. Local R. 13.165 (2013). “Approximately two months are allocated for the arbitration process.” See Or. Unif. Trial Ct. R. 13.160(4); Multnomah Cnty. Local R. 13.165.

  207. [207]. Summary Statement from Doug Bray, supra note 205.

  208. [208]. Multnomah Cnty. Local R. 7.075.

  209. [209]. Id. Although judicial settlement conferences were traditionally free of charge and now cost only $50 per party, neither mediation nor arbitration is offered on a free or subsidized basis. See Or. Judicial Dep’t, Multnomah Circuit Court Fee Schedule (2009), available at http://courts.oregon.gov/Multnomah/docs/CourtFees/MCCCFeeSchedule_10.2009.pdf. Mandatory arbitration programs in the Oregon Court have an hourly fee of $125.00, with a maximum cost of $1,000. The court also provides a list of court-approved mediators with fees ranging from $75 to $300 per hour as well as some daily rates. In the Matter of the Compensation of Arbitrators Appointed Under UTCR Chapter 13, Order No. 070100000 (Or. Cir. Ct. Jan. 18, 2007), available at http://courts.oregon.gov/Multnomah/docs/Civil
    Court/CourtMandatoryArbitrationPrograms_ArbitrationFees.pdf; Fourth Judicial Dist., Or. Judicial Dep’t, Court Annexed Civil Mediators (2013), available at http://courts.
    oregon.gov/Multnomah/docs/civilcourt/civil_annexed/all_civil.pdf.

  210. [210]. See E-mail from Doug Bray to Shannon Clawson, supra note 200.

  211. [211]. Oregon Unif. Trial Ct. R. 7.030(2) (“The criteria used for designation as a ‘complex case’ may include, but shall not be limited to, the following: the number of parties involved, the complexity of the legal issues, the expected extent and difficulty of discovery, and the anticipated length of trial.”).

  212. [212]. Id. at 13.050.

  213. [213]. Multnomah Ctny. Local R. 12.025(2).

  214. [214]. Id.

*

Professor of Law and Martin Luther King Jr. Scholar, University of California, Davis, School of Law. J.D., Stanford Law School; Ph.D. (Psychology), Stanford University; M.S. (Psychology), Yale University; B.S., Yale University. Correspondence should be addressed to Dr. Donna Shestowsky, University of California, Davis, School of Law, 400 Mrak Hall Drive, Davis, CA 95616.