• What did you expect from filing for divorce or modification, from the court, from your attorney?
  
• Did the legal process satisfy your expectations?
  
• How did your case get resolved? Was there a turning point in the process that lead to resolution?
  
• If you could do it over knowing what you know now, would you do things differently?
  

One variable that has greatly interested us has been the whether the passage of time makes a case “ripe” or “stale” for the purposes of reaching a mediated settlement. Opinions vary as to the proper time to mediate. One commonly hears that cases are “not ready” for settlement discussions because the parties have not had the opportunity to conduct sufficient discovery to evaluate a case for settlement purposes.

To explore these ideas, we determined for each case in our sample how many months elapsed from time the case was filed to the time the case was mediated.

In Cobb County Superior Court’s ADR Program, cases are referred to mandatory mediation a certain amount of time after the answer is filed, typically one or two months. For a variety of reasons, including scheduling conflicts and delays, mediation may not actually take place for many months following referral, as the following graphic illustrates:

Enlarge Graphic

We determined the mediation settlement rate during each interval to see whether the passage of a certain amount of time between cases filing and mediation correlated with settlement or impasse.

Enlarge Graphic

Some reported mediations took place 0 – 3 months after case filing, before one would expect a mandatory mediation referral, and those cases appear to settle at a relatively high rate. There appears to an interesting correlation between months in which a relatively large volume of cases are mediated (e.g. months 6 and 9) and a relative drop in settlement rates in those months. This may suggest that “pushing” cases to mediation causes a drop in settlement rates.

We were also interested to see whether the passage of time had a different effect on domestic relations mediation than civil/damages cases. One may hypothesize that the opportunity to learn about a case over time is more significant in civil/damages cases where the parties are relative strangers than in domestic relations cases.

Enlarge Graphic

Enlarge Graphic

The passage of time does not appear to have a noticeable effect on the settlement rate of civil/damages mediations, but a larger data sample of such cases may be helpful to determine whether there is any correlation between the passage of time and probability of mediated settlement. The settlement rate appears to “spike” in mediations conducted 12 months after the case was filed. We may speculate that mediating near the year anniversary of a case may have some psychological or contractual importance for the parties, but do not have a solid hypothesis for this feature of the data sample.

Our regression analysis of the passage of time against settlement rates also did not reveal a strong correlation. According to regression analysis of the data sample, from a settlement rate of 61.6% at the time of filing, the mediated settlement rate declined approximately one-half percent for each month that elapsed to mediation. This linear regression does not fit a lot of the data, suggesting there are more variables in mediated settlements than the passage of time.

The data on passage of time suggests that there is no “right time” to mediate a case. Perhaps the right time to mediate is the time when mediation is agreeable to the parties. There is some evidence that pushing parties to mediate to comply with administrative rules may decrease settlement rates. The evidence may also suggest that the opportunity to conduct discovery and file pre-trial motions does not significantly alter, and may actually decrease, the parties willingness to settle at mediation. This would suggest that discovery and motions practice may more often be employed to justify a party’s settlement position than allow parties to evaluate the strengths and weaknesses of their case for settlement purposes.

To produce this graphic, we plotted the settlement rates of cases mediated in Georgia’s large Superior Court mediation programs. This includes programs where more than 100 cases per year were mediated from 1997 to 2005. This graphic shows the percentage of cases reaching any type of settlement (full or partial) in mediations administered by the superior court programs represented here.

Click to Enlarge Graphic

This table is limited to superior court ADR programs that administer more than 100 cases per year. Some data was not available (DeKalb 1997, Gwinnett 1998, Coweta 2004 excluded due to likely recording error).

Sadly, no data for 2006 and 2007 is available from GODR. The office reports its work has been limited by staff and budget cuts.

The settlement rates of Superior Court Mediation Programs are fairly consistent over time. The Southern JAD reports a relatively high rate of settlement through mediation whereas the Cobb Superior Court ADR Program reports a relatively low rate of settlement over the time period reflected here.

A number of variables may affect the settlement rates achieved in the various programs represented here. We might expect that these programs would mediate the same types of cases because we have isolated Superior Court programs. Other variables that may account for the various settlement rates achieved include

• Timing of Referral to the ADR Process;
• Regulation of Who May Mediate Court-Connected Cases;
• Demographics of the Population Served;
• Party Involvement in Selecting Mediator, Mediation Location; and
• Who Pays the Mediator

Our hypothesis is that the more a court program exercises coercive control of the mediation process, the less likely cases mediated through that program will result in settlement.

Although no less than the Georgia Supreme Court has created a Commission to continuously evaluate the performance of court-connected ADR programs, I have received scalding criticism from court administrators for evaluating how well their mediation programs have actually performed in terms of settling cases and representing that court programs should be evaluated for the benefit of litigants and taxpayers who fund these programs.

This graphic has been published on our web site, www.centerforlegalsolutions.org, for some time, but I wanted to post it separately to provide some additional commentary on it and provide readers an opportunity to share their comments on it.

The graphic below shows the percentage of cases reaching any type of settlement (full or partial) in mediations administered by court-connected mediation programs sorted by type of court. In other words, the produce this graph, all superior court mediation programs were grouped together, all state court programs grouped together, etc. We hope that presenting this data in user-friendly form will help those interested in this field.

Click to Enlarge Graphic

The State Court Programs settlement rate (pink line) is dramatically impacted in 2005 by the addition of cases from the 9th JAD State Court Medation Program. In 2005, the first year of reported state court data from 9th JAD, there were 317 full or partial settlement in 327 cases (a 97% settlement rate).

Taking note of the exceptional state court results from 9th JAD in 2005, the overall settlement rates by type of court appear to remain fairly stable over time. Juvenile and Magistrate Court cases appear to settle through mediation at relatively high rates. The settlement rates in Probate and Superior Court cases appear in a middle range. The relatively small number of Probate Court cases being mediated accounts for the apparent variability of the settlement rate of Probate Court cases (the yellow line). Settlement rates are relatively low in State Court Mediation Programs.

The types of courts represented here have different (sometimes overlapping) jurisdictions. This data indicates that mediation produces settlements at higher rates in some types of courts than others. This supports the hypothesis that some types of cases are more suited to settlement at mediation than others.

This graphic has been available on our web site, www.centerforlegalsolutions.org, for some time, but we wanted to isolate our discussion of the court_type variable in this post to give readers an opportunity to comment.

We identified the gender of the 122 individuals who mediated these cases in order to compare the overall settlement rate of male mediators compared to the overall settlement rate of male mediators. Our findings can be summarized.

Female Mediators	300 Cases	62.0% Settlement Rate
Male Mediators	278 Cases	42.4% Settlement Rate

Female mediators settled cases in our sample settled cases at approximately 20% higher rate than male mediators. The overall settlement rate was 53%.

To further illustrate the data, we organized our chart of “popular mediators” according to the settlement rates of the 35 most frequently used mediators in our data sample. We then colored each bar to represent the gender of the mediator who achieved the settlement rate represented.

View Enlarged Graphic

Examining the graphical representation of mediator gender and settlement rates, one can see that there are male mediators who settle cases at higher than average rates, as well as female mediators who settle cases are lower than average rates. Nevertheless, it appears that most of the popular mediators who settle cases at higher than average rates are women, while the majority of popular mediators who settle cases at lower than average rates are men.

Some may object to this “battle of the sexes” analysis on the grounds that men and women should be treated as equals. Based on our data, however, male and female mediators are not statistically equal with respect to the rate at which they settle cases. Whether this “good” or “bad” is more a matter of philosophy than statistics.

In her book In a Different Voice, Carol Gilligan described how men and women think about moral conflicts differently. Her research suggests that men tend to consider conflict in terms of rights while women generally view conflicts in terms of dynamic relationships. Accordingly, a “female” approach to conflict resolution may be better suited to the process of facilitating mediated settlements than a “male” approach to conflict.

Because many cases in our data sample involve family law conflicts between parties with long-term relationships, as opposed to conflicts between relative strangers, settling these cases may require mediators to appreciate the relationships of the parties involved.

As we continue to study mediated cases and increase our data sample, it will be interesting to compare the settlement rates of male and female mediators in domestic relations cases against other types of cases, particularly lawsuits for damages.

There are 10 active Cobb County Superior Court judges. Our data sample of mediated cases was sorted by the judge assigned to the case to determine the rate at which their cases settled in mediation. The results can be illustrated graphically.

Enlarge Graphic

Based on our data, the judge presiding over the case does have an influence on the settlement rate. Cases in our data sample settled at an average rate of 53% in our data sample.

In the Cobb Superior Court ADR Program, cases are automatically referred to a mandatory ADR process based on the elapse of a certain amount of time (usually 1 – 2 months) after the complaint is answered. Because the process is largely administrative, it is unlikely a particular judges “positive” or “negative” sentiments about mediation explain the varying settlement rates when their cases are mediated in this program. It is more likely that these judges indirectly influence the rate at which their cases settle in mediation through their case management. For example, it might be expected that how pre-trial motions are handled before mediation and the expectation of an imminent trial play a role in whether mediation parties will be able and willing to reach settlement at mediation.

It should also be noted that Judge Kell (Judge #10) was appointed to the Cobb Superior Court after the mediation sessions that compose our data sample occurred. Only active cases where mediation resulted in impasse, partial settlement or continuation of settlement were re-assigned to Judge Kell when he took the bench in late 2007.

Our empirical analysis of mediated cases doesn’t offer us a crystal ball to discern whether a mediated settlement would have occurred naturally, without the involvement of a mediator, but has disclosed an interesting feature of out-of-court settlement rates in court-connected ADR programs in Georgia.

Generally, when a case is referred to a mandatory ADR process, the parties have some time make arrangement or prepare for a mediation session arranged by the court. This provides the parties some opportunity to resolve the matter on their own, prior to mediation. Between 1997 and 2005, the Georgia Office of Dispute Resolution recorded the number of cases that where referred to mandatory, court-connected ADR but were resolved prior to mediation. We calculated the percentage of referred cases that were resolved prior to mediation in different types of court over the time period of available data. A graphical depiction of the data reveals several upward trends:

Enlarge Graphic

It appears that in state, superior and juvenile courts an increasing percentage of cases referred to mediation settle on their own before mediation actually takes place. Apparently, the natural settlement rate is not a natural rate, but rather responsive to some variables. One might hypothesize that an increasing percentage of parties are settling their cases out of court to avoid the time and expense of court-ordered mediation.

While courts should encourage voluntary settlements, one has to be concerned that the burden of court-ordered mediation may be a deterrent to pursuing justice under the law. If courts were simply to impose a $1,000 fine on all parties who refused to settle their case within a certain amount of time, such a rule would raise serious due process concerns but may be a more efficient and direct method of promoting settlements than ordering parties to mediation sessions that they are increasingly reluctant to endure.

Georgia’s Superior Courts are courts of general jurisdiction. They hear all types of civil cases and wide variety of cases are referred to mediation programs connected with the Superior Courts.

To determine whether certain types of cases more likely to settle through mediation than others, our data sample was sorted by case types. The case types are assigned by the Clerk’s office. We then compared the settlement rate of each type of case that completed a mediation session.

Enlarge Graphic

Based on our sample of 578 mediated cases, which we believe to be representative, some types of cases are more likely to settle during mediation than others. Cases concerning specific domestic relations issues such as separate maintenance, paternity, visitation and child support appear to settle at relatively high rates through court-connected mediation while cases for damages and civil complaints settle at relatively low rates through mediation.

Enlarging our data set may help us calculate the suitability of more exotic forms of action, not listed here, for settlement in mediation.

These findings should be useful to attorneys who would like to advise their clients on the value of mediation for their particular dispute. Mediation outcomes can not be predicted with certainty, but studying the data may support an educated guess. These finding would also be useful to court administrators who wanted to target resources where they may do the most good.

It is perhaps significant that the cases in this sample are generally referred to mediation at a particular time (one or two months) after the answer (or answers) are filed. It may be that certain cases are better suited to mediation early in the life of a case than others. Parties in cases for damages and civil complaints may have a greater need for fact finding than parties in domestic relations matters which makes early mediation more appropriate for the latter than the former. Multiple-variable analysis and exploration of the time factors involved should be helpful.

These findings are based on the reported outcomes of 578 mediation sessions conducted through the Cobb County Superior Court ADR Program. We have utilized the Superior Court Clerk’s web site to mine data from Mediation Reports covering approximately 9 months in 2006 and 2007. These Mediation Reports should be a representative sample, though we are organizing more data to make additional findings.

One question we seek to answer is whether individuals who are frequently selected to mediate cases settle cases at a higher rate than individuals who are selected less frequently to mediate cases. In Cobb County Superior Court, parties are ordered to mediate, but they have an opportunity, usually 13 days, to select their mediator. If the parties cannot agree upon a mediator, the Court Program assigns a mediator to their case.

In our sample, we found that some individuals mediate far more frequently than others. Overall, 121 different individuals served as mediators. 11 individuals mediated more than 8 cases in our sample; 36 mediated between 3 and 7 cases; 74 mediated 1 or 2 cases. We calculated the settlement rates for the 35 most popular mediators in our sample (those who mediated 4 or more cases). Cases that resulted in partial settlement or continuation were considered 50% successful, reflecting a middle point between full settlement and impasse.

The results can be illustrated graphically:

Enlarge Chart

The overall settlement rate in our sample was 53%. As one can see from the graphic above, some frequently used mediators settled cases at below average rates while some less frequently used mediators actually settled cases at relatively high rates.

The data allowed us to compare three groups of mediators.

Category:	Number of Cases:	Settlement Rate:
8+ Cases	329	57.0%
3 – 7 Cases	145	53.4%
1 or 2 Cases	104	40.4%
All Mediators	578	53.1%

Based on the data, frequently used mediators do settle cases at a higher rate than average. In our sample, however, we found significant variability among individual mediators and note that difference in settlement rates between mediators who handled 8+ cases and those who mediated between 3 and 7 cases is not as dramatic as one might expect.

Finally, our sample includes cases were the parties selected their mediator as well as cases were a mediator was assigned. It is likely that those who mediated less frequently had a higher proportion of assigned cases than those who mediated more frequently which may account for some of the difference in settlement rates observed in our data.

Parties make decisions to further their self interest throughout the litigation process. If engaging in mediation can help the parties, presumably the parties will mediate without being ordered to mediate. Parties engage in direct settlement negotiations and mediate voluntarily all the time.

Surely ordering parties to mediate creates more work for mediators, engages court administrators in court-connected programs and reduces judicial workload, but these are not good reasons for ordering parties to mediate.

Mediation is supposed to be a process that allows parties to exercise self-determination in order to reach a voluntary settlement as an alternative to litigation. Ordering parties to mediate under there threat of court sanction may be fundamentally at odds with the core values of mediation. To the extent that parties are able to exercise self-determination and decisions about when, how, where, with whom and whether to mediate, I believe their chances of reaching settlement increase. The more coercion is employed to force parties to mediate, the less chance the parties have of reaching settlement.

When mediation was a new process that parties could not fully understand or appreciate, courts may have been justified in ordering mediation to overcome the parties lack of knowledge. Mediation is now no longer an unknown or experimental process. If parties are so irrational that they can’t weigh the pros and cons of mediation for themselves, how can they be expected to weigh the pros and cons of settlement offers during mediation? While ordering parties to mediate may serve some special interests, I would question whether it is a good general rule.