Assessing Empowerment in Divorce Mediation

The practice of divorce mediation is guided by a number of assumptions:

1. Conflict is a natural product of human relations and thus represents an opportunity for positive change (Haynes and Haynes 1989);

2. Mediation offers an appropriate forum to resolve issues facing divorcing couples, including custody, child support, spousal support, and property division (Folberg 1983; Haynes and Haynes 1989; Haynes 1994);

3. Mediation is a goal‐oriented, problem‐solving process (Folberg 1983; Brown 1985; Haynes 1994);

4. The outcome of mediation is the responsibility of the parties, not of the mediator (Folberg 1983; Brown 1985; Haynes and Haynes 1989; Haynes 1994); and

5. The mediator controls the mediation process as the guiding third‐party neutral (Folberg 1983; Haynes and Haynes 1989; Haynes 1994; Emery 1995).

These assumptions appear to be fundamental to the practice of divorce mediation; indeed, the legitimacy of mediation as an alternative to adjudication in these cases is largely predicated upon these assumptions. While the literature is sprinkled with debate over such matters (Harrington 1985), proponents of mediation argue that the process of mediation is important in itself, regardless of outcome, because it enables people to have more control over the critical decisions that affect their lives. In other words, mediation empowers people during a difficult time when the stakes are high, not only for their own futures but also those of their children.

Julian Rappaport (1984: 3) defined empowerment as a process by which people “gain mastery over their lives” and argued that the process of empowerment and its outcomes are difficult to delineate because each is dependent upon unique persons in unique circumstances. Empowerment is not merely the capacity to achieve control over one’s aims but also the ability to actually achieve those aims (Rappaport 1984).

More precisely, empowerment refers to the process of achieving power (Staples 1990; Robbins, Chatterjee, and Canda 1998), the ability to exercise that power (Staples 1990), and the exertion of real control in one’s own interests (Staples 1990; Robbins, Chatterjee, and Canda 1998). In sum, empowerment involves developing the ability to both acquire the requisite social resources and to facilitate the use of those resources in order to influence a desired outcome. For mediators who advocate a style of mediation known as the transformative approach (Bush and Folger 1994; Etcheson 1999), empowerment is a key component of a successful mediation process because it provides the individual, regardless of the outcome, the strength and capacity to better handle life’s current and future challenges.

Research in the area of divorce mediation specifically has found that clients who mediated would recommend mediation to others (Pearson and Thoennes 1984; Thoennes and Pearson 1985), felt “heard” in the mediation (Kelly 2004), and were pleased with mediation generally (Pearson and Thoennes 1985; Kressel et al. 1994; Kelly 2004) even if a settlement was not reached (Pearson and Thoennes 1985). Researchers have also found that satisfaction of both parties in a dispute was more balanced in mediated than in adjudicated cases (Emery and Wyer 1987) and that clients in problem‐solving mediations were pleased with the mediation and reported it had positive effects upon the coparental relationship (Kressel et al. 1994).

While participants may be generally pleased with mediation, it does not necessarily mean they are empowered in it. Indeed, Robert Emery, Sheila Matthews, and Katherine Kitzmann (1994) found that fathers who mediated custody issues reported higher feelings of control over decisions than those who litigated, but they found no differences in feelings of control between mothers who mediated versus those who litigated. One possible interpretation of these findings is that fathers may feel more empowered by mediation than litigation but mothers do not feel empowered by either process. Feeling empowered, however, is only one aspect of empowerment; another essential feature of empowerment is a change in actual material outcome. That empowerment is possible and can be critical to the success of mediation is largely agreed upon; the challenge for empowerment research is to develop ways to effectively measure this phenomenon (Speer 2000).

What does empowerment look like and how can it be measured in both the process and outcomes of divorce mediation? Based on interviews with family mediators, Jo Daugherty Bailey (2007) reported that it was possible to examine the sections of divorce decrees that specify child custody/visitation arrangements and assess that either empowerment or disempowerment had occurred in the mediation. In Texas, a legislatively defined Standard Possession Order (SPO) detailed in the Texas Family Code, §153.312 (http://tlo2.tlc.state.tx.us/statutes/docs/FA/content/pdf/fa.005.00.000153.00.pdf), outlines the default arrangements for custody and visitations, arrangements that can be modified in litigation or mediation to better fit a family’s particular circumstances. The mediators interviewed reported that a custody/visitation arrangement that was more “skewed” (i.e., offering one parent significantly less time with the child) than the standard order would indicate that one party, the one “cut out” of the child’s life, had been disempowered in the mediation process. A decree that arranges custody/visitation more equally than the standard order would be a sign of couple empowerment.

In a previous study, Bailey and Susan Robbins (2005) examined the disposition of child custody in 120 randomly selected divorce cases in Texas and found that couples who mediated were significantly more likely to make changes to the SPO than couples who did not mediate and that the number of changes in the mediation group were significantly greater. Specifically, the probability of making a change was three times greater for mediated cases than for nonmediated. That study, however, left open the possibility that changes in the SPO might have been disempowering rather than empowering. Research has shown that a child’s continual, stable contact with both parents after divorce indicates parental cooperation, while infrequent, unstable visits point to parental conflict (Ahrons 1994) and that stable contact is associated with better outcomes for the child ((McDonough and Bartha 1999). If the mediated changes to the SPO studied by Bailey and Robbins (2005) decreased the contact of the noncustodial parent with the child, mediation, it could be argued, might be doing more harm than good.

To test whether the increased changes in the SPO were more or less empowering, we reexamined the 120 randomly selected mediated divorce cases from the previous Bailey and Robbins (2005) study. Many of the decrees incorporated the SPO word‐for‐word, but in some cases, a great many changes were made to the SPO. It is often not immediately apparent that a parent has been “cut out” of his or her child’s life or, on the other hand, that the possession order in a specific case is more “equal” than the SPO, so we developed a classification matrix and applied it to the cases.

A divorce decree can affect the parent–child contact available to the noncustodial parent in two ways and in two directions: the continuity of contact with the parent over the course of the year may be impeded or facilitated, and/or the extent of each visit may be reduced or expanded. Thus, several types of changes in the SPO may indicate either a skewed agreement associated with disempowerment or a more equal agreement associated with empowerment. Any case in which more than half of the seven visitation categories (weekends, Wednesdays,¹ spring school vacation, major holidays, child’s birthday, Father’s Day/Mother’s Day, and summertime) had been altered offered the possibility of affecting both the continuity of visitation and the extent of visitation.

Because these visitation categories vary in the actual duration of the visit, a change in specific categories could disproportionately affect the overall amount of visitation, which complicates our analysis. Three categories (weekends, Wednesdays, and summer vacations) have a disproportionate impact on the continuity of contact and the amount of time spent with a child. These changes must be considered in terms of the interactions with each other. If a change is made to only one of these categories, the impact might be manageable because the noncustodial parent could recoup the lost time through the other scheduled visitations. For example, if the weekend visitations were excluded, the parent would still be able to visit with the child on Wednesdays, facilitating continuity of contact. If visitation on both weekends and Wednesdays were dropped, however, the parent would have to wait until a holiday for visitation. Restrictions on summer visitations have a similar likelihood of impeding both continuity and the amount of contact that the noncustodial parent and the child have with each other.

Given the potential impact of these visitation periods, we included in our analysis only cases in which both the weekend category and the Wednesday category had been changed and cases in which both the weekend category and the summer category had been changed. The three categories of cases that we then analyzed were:

• four or more visitation categories changed (fourteen of 120 cases);

• weekend and Wednesdays changed (twenty‐three of 120 cases); and

• weekend and summer changed (twenty of 120 cases).

We found considerable overlap in the cases that fell into more than one of the above categories, resulting in twenty‐one mediated and six nonmediated cases for analysis.

We examined each of the twenty‐seven cases to summarize and describe the nature of the changes made within each category and organized the results into a matrix. We then compared the description summaries across cases. Four types of changes to the SPO emerged from the data: minimal, moderate, extensive, and radical.

The minimal change group consisted of three of the twenty‐one mediated cases and all six of the litigated cases. Most of these cases simply changed the beginning or ending time for the visit, either slightly increasing or decreasing the time. These time changes were usually made in a way that benefited each party. The small number of mediated cases in this group suggests that changes made to visitation orders in mediation seem more likely to be more significant than changes made to visitation orders in litigation. The fact that no litigated cases included more than minor changes indicates that changes to the SPO that occur because of litigation are less common (Bailey and Robbins 2005) and more minor.

Ten cases fell into the moderate change category, making this the largest group in the study. Changes that were made to the SPO in every one of these cases expanded the visitation time for the noncustodial parent. For example, in several cases, the arrangements for weekends and Wednesdays were altered so that each parent had the child for half of the week during the school year. While the changes made in these cases were always in the direction of more equal time spent with each parent, the custodial parentstill maintained a majority of custodial time with the child in cases in this category.

For a third group, extensive change cases (seven cases), the SPO was changed to give an absolutely equal distribution of possession time between parents, reflective of a true coparenting joint custody arrangement. (Two additional mediated cases almost could be included in this category but were placed in the moderate change category because the summer vacation schedule was not exactly equal.) All these cases resulting from mediation were a dramatic departure from the SPO, a result that was never reached in any of the litigated cases.

In one case, the sole radical change case, the SPO was changed in the opposite direction: the noncustodial parent’s visitation was limited to two days a week throughout the year. No special provisions were made for holidays, birthdays, summers, or Father’s Day. At first glance, this fits the mediators’ descriptions (Bailey 2007) of a case in which empowerment did not occur for both parties because the noncustodial parent appears to be disempowered — under such an arrangement, the parent–child contact would be continual but not extensive. We further explored the field notes made on the case and discovered that the parties had three‐year‐old twins and that the decree stated that when the children turned five, they would revert to the SPO. This provides an alternative explanation to disempowerment in this case. The parents may have felt that the children were too young for extensive stays away from their primary residence but that in two years they would adjust better to such a transition. Rewriting the entire possession order to reflect their interpretation of their children’s needs would seem to indicate a high degree of empowerment for both parties.

Regarding legal representation, in seventeen of the cases, both parties were represented by an attorney. In just four cases was only one party represented (the plaintiff in each case); two of those cases were mediated and fell into the moderate change category, and two were litigated and fell into the minimal change category. Joint managing conservatorship was awarded in all but one of the twenty‐one cases. In sixteen of the cases (five of these were litigated and eleven were mediated), the children primarily resided with the mother, in five cases (one was litigated) the father was the primary custodial parent, and in six (all mediated) there was no primary residence indicated in the record.

In summary, the hypothesis that changes to the SPO in mediated agreements will restrict visitation time or otherwise disempower the noncustodial parent (Bailey 2007) was not supported. The nature and types of changes in twenty of the twenty‐one mediated cases in no way limited the time or extent of visitation for the noncustodial parent; in fact, the majority of mediated cases expanded the amount of visitation, resulting in a more equitable distribution of child contact time between the parties.

The findings from this study enrich our understanding of issues of fairness and empowerment in divorce mediation. Some of the mediation literature advises mediators to be evaluative during mediations, to watch for one party exercising undue power over the other, and to take “reality checks” with each party regarding the allowances and concessions they are making to verify that they understand the impacts of what they have agreed to (Haynes and Haynes 1989; Emery 1994). Because the circumstances of each mediation are so unique, however, prescriptions for conducting ongoing assessments of empowerment and disempowerment in mediations are problematic. Indeed, Connie Beck and Bruce Sales (2000) noted that no research offering an operationalization of power imbalances in mediation had yet been published.

This study offers such an operationalization, as well as a mechanism for evaluating empowerment and disempowerment during mediation based on the presence of an unequal or skewed SPO agreement. By assessing the skewness in the agreement and by openly discussing it with the parties, a mediator who holds an empowerment view of mediation could be better assured that both parties have been provided as much opportunity for empowerment as possible.

These findings have implications for litigated divorces, as well. Standard Possession Orders are useful in litigation because they offer a benchmark for attorneys. But it is possible that, driven by societal expectations, courts, attorneys, and parties will grant more validity to these templates than they deserve. Because there is no evidence that the SPO is any more or less applicable to mediated than to litigated cases, the finding that SPOs are much less often modified during litigated than mediated cases suggests that all parties in the process grant too much importance to SPOs and support changes too infrequently.

These findings also have policy implications for family law. The current SPO in Texas grants visitation to the noncustodial parent the first, third, and fifth weekends; Thursday of each week (two hours in the evening dedicated or overnight if allowed); alternate major holidays; and a portion of the summer vacation. Not counting alternating nonsummer holidays including spring school vacation, birthdays, etc., the noncustodial parent receives for the year about sixty contact days on weekends during the school term, thirty‐nine Thursday visits during the school year, and thirty days in the summer, totaling 129 days with alternative yearly holidays not factored. In terms of the welfare of the child, is this enough or too much? Of course, each case is different, and that is precisely the greatest advantage of mediated divorce, that mediation more flexibly accommodates differences and individual family’s unique needs and circumstances.

Nevertheless, the details of the SPO, which serve as a benchmark in both mediation and litigation, are extremely important. The results of our study suggest that the SPO in Texas grants too little time to the noncustodial parent because when given a chance to create more equitable visitation arrangements, the majority of parents in our study chose to do so. It seems a reasonable assumption that a mediated agreement freely entered into by both parties and, in Texas at least, in the presence of their attorneys, represents the best division of time with the child in that particular case in the opinion of the parents involved, who are more likely to know the best interests of themselves and of their child than the Texas legislature.

The findings of this study are clear: in mediated cases in which the SPO was changed, it almost always (20 of 21 cases) was changed in the direction of granting more time to the noncustodial parent than the SPO called for. This suggests the possibility that Texas and other states that similarly restrict noncustodial time may need to re‐evaluate their standard orders.

The concept of skewness as an indication of disempowerment seemed to work well in analyzing divorce mediation, but it may be less useful in analyzing other types of disputes. In addition, in looking specifically at divorce cases, this sample consisted of a relatively small number of Texas cases, limiting the ability to generalize the findings to other states. Currently, nine states have moved to a presumption in favor of joint custody (American Bar Association 2008). Further research in these states in particular (California, Connecticut, District of Columbia, Idaho, Minnesota, New Hampshire, New Mexico, and Oregon) regarding the causes and conditions that led to a reliance on more equitable, less skewed custody arrangements could provide additional support for the findings presented here.

Also, Ohio relies on a standard order of visitation that is similar to the Texas order, and although much effort has been made in recent years to tailor the Ohio order to meet the age‐appropriate developmental needs of children, most counties, when they had the option, failed to utilize the new orders in favor of the old (Dember and Fliman 2005). This failure of policy makers to adopt a policy that experts agree would be in the best interests of children seems to support the process findings reported here: that as participants themselves have the power to make changes to any standard or imposed order, they are more likely to do so, and thereby demonstrate their level of empowerment in the process. Further research is needed to support this implication.

Most importantly, this study, an objective analysis of the distribution of outcomes of mediation, did not examine the perceptions of the disputants. Some studies have indicated that mediating couples report feeling more involved in the process and in greater control of their outcomes thannonmediating couples (Kelly 2004), but others have not (Beck and Sales 2000).

In her critique of mandatory mediation, Trina Grillo (1991) argued that women have more to lose in mediation in that mediation norms may exacerbate gender inequalities. Others (Bryan 1992; Meyers 1996) have echoed similar concerns. Thus, shared custody may reflect a woman’s “loss” of some social status and power. One of Grillo’s central concerns is the tendency to exclude attorneys, whose function is to mindfully protect their client’s interests, from mediation. Given the propensity for shared parenting we found in the cases we studied, one could argue that mediation that results in coparenting arrangements generally does indeed reflect maternal disempowerment. But that argument is challenged by the fact that, in nineteen of the twenty‐one mediated cases we studied, attorneys represented both parties. Nonetheless, Grillo’s argument suggests that family law mediators should be cautious of the possibility of disempowerment, especially in cases in which one or both of the parties do not have legal representation.

Further research that incorporates, in both mediated and litigated cases, the disputants’ own assessments of their empowerment together with objective measurements of empowerment, would be very useful.