Triangulation of Salient Studies to Date on Trust‐Building in Mediation

There is a consensus in the literature on the importance of trust‐building between mediators and disputing parties (McCarthy 1985; Poitras 2009; Stimec and Poitras 2009). Various studies have demonstrated that building a relationship of trust between mediators and disputants is a key factor in a mediation’s success. In one study, over 75% of a survey’s respondent mediators indicated that success in mediation is associated with building a rapport; that is, “a relationship of understanding, empathy and trust” (Goldberg 2005: 366). The main benefits of trust‐building are shown in Figure One.

As shown above, the parties’ commitment to the mediation is an intermediate step between building a rapport and the likelihood that the mediation will result in an agreement between the parties. Recognizing the importance of the parties’ commitment, Jean Poitras and Robert Bowen (2002) researched the following question: If the parties’ commitment to the process is an intermediate step between rapport‐building and the possibility of reaching agreements, has sufficient research been conducted on how to initiate the mediation process in order to foster a sense of commitment by the parties? This study sought to answer this question.

Table One contains a bibliographical review of the factors that increase the generation of trust between mediators and disputing parties. It is limited to factors that were cited by five or more publications in the literature that we reviewed. (We reviewed fifty articles selected from specialized databases.) According to Gulati (1995), these factors can be applied to any type of mediation.

The degree to which trust‐building is influenced by the factors listed above will depend on the stage in the mediation during which they are perceived. In other words, the length of time that passes before a disputant believes the mediator to be credible or impartial will affect how strongly the disputant’s perception of such credibility or impartiality influences trust‐building. Trust‐building during the preliminary stage of the mediation process stabilizes the relationship of the participants (McKnight, Cummings, and Chervany 1998; McKnight and Chervany 2006). After a certain point, these factors are less likely to increase the possibility of an agreement between the parties than if they were recognized during the mediation’s preliminary stage (Davis and Gadlin 1988; McKnight, Cummings, and Chervany 1998; Stimec and Poitras 2009).

Our research tests theories on the factors that increase trust‐building in mediation. We conducted this research in response to a request by Stephen Goldberg and Margaret Shaw (2007) for studies conducted in North America to be triangulated in other countries where mediation is taking off as an alternative tool in dispute resolution. With this in mind, the study conducted by Poitras (2009) in Montreal (Canada) (n = 105) was triangulated in the Balearic Islands (Spain) (n = 54). Our hypothesis was that despite their different cultural contexts and their applications to different types of mediation, we would be able to compare the results of Poitras’ study and our study regarding the factors that increase trust‐building in mediation.

Our study required the collaboration of the Government of the Balearic Islands’ Regional Department for Social Services and Cooperation as well as the mediators in its external Family Mediation Service. In 2015, 31.34% of the mediations conducted by the Family Mediation Service resulted in a partial or total agreement between the disputants. Most of the disputes involved divorcing couples with at least one child.

The non‐probabilistic intentional sample was made up of all the users of the Government of the Balearic Islands’ Family Mediation Service during the twelve months from March 1, 2017 to March 1, 2018 (n = 54). In sociodemographic terms, the users were evenly distributed between males and females (46.3% males versus 53.7% females); most of them were in the 36‐ to 45‐year‐old age bracket (48.1%); and 43.1% of them lived in a place with a population of 5,000 to 40,000 inhabitants. Most people in the sample were separated from their spouse (64.1%); 90.7% of them stated that they had one to two children; and 70.4% of them said that they were employed. As for their level of education, 30.8% of them said that they had undertaken university studies and, when they were asked whether they had a previous positive experience with other professional support services, 50% stated that they had visited a psychologist.

At our request, the Family Mediation Service’s mediators informed its clients of the possibility of taking part in the study and introduced participants to the researcher. Since there is a consensus in the relevant literature that trust between mediators and disputing parties is built during the beginning of the mediation process (Davis and Gadlin 1988; Landau and Landau 1997; McKnight, Cummings, and Chervany 1998; Butler 1999; McKnight and Chervany 2006; Stimec and Poitras 2009), we were particularly interested in this period of time. Therefore, the mediators asked participants to fill in a questionnaire at the end of their third joint session; the entire mediation process generally lasted for six to eight sessions.

The questionnaire was comprised of closed questions assessed on an ordinal polytomous scale. Based on prior research findings—including the work of the researchers identified in Table One—the questions incorporated the factors that have been shown to increase the generation of trust (Riera Adrover 2017). The questionnaire was analyzed on the basis of ten attributes of the mediator: mastery (experience, knowledge of the case, self‐confidence), explaining the mediation process, warmth and consideration, chemistry, credibility and reputation, experience with the type of dispute, skills for the intervention, mediation principles, goodwill and empathy, and skills for building a therapeutic alliance. It contained a total of sixty‐five questions, answered on a 5‐point Likert‐type scale. The questionnaire is available from the authors (Riera Adrover and Casado 2018).

The reliability statistics gave a Cronbach’s alpha of 0.866 for the dependent variable; 0.831 for the mastery part; 0.730 for the explaining the mediation process part; 0.922 for the warmth and consideration part; 0.852 for the chemistry part; 0.783 for the mediator’s credibility and reputation part; 0.716 for the experience with the type of dispute part; 0.789 for the skills for the intervention part; 0.780 for the mediation principles part; 0.797 for the goodwill and empathy part; and 0.906 for the skills for building a therapeutic alliance part. The minimum reliability value is 0.70 and so values of over 0.70 are acceptable.

The users also filled in another questionnaire containing closed questions assessed on an ordinal polytomous scale, made up of three items aimed at measuring the general trust that was built up between the mediator and the disputing parties. This was also answered on a 5‐point Likert‐type scale. This questionnaire was an adaptation of the scale from the original study by Jean Poitras (2009).

A statistical analysis of Pearson’s correlation coefficient (IBM* SPSS* Statistics Version 25) was conducted between the results of our instrument (CONMEF) and the results of the measurement scale of trust between mediators and disputing parties (Poitras 2009).

The main difference between this study and the one by Poitras (2009) is that Poitras’ study was based on a qualitative methodology aimed at analyzing the parties’ perceptions of the factors that increase trust‐building in employment mediation, while this study is based on a quantitative methodology aimed at exploring the parties’ perceptions of the factors that increase trust‐building in family mediation.

However, by applying the “Measurement Scale of Trust between Mediators and Disputing Parties,” the factors from the qualitative method used in Poitras’ study (2009) may be compared with the factors from this study’s quantitative method. Through the use of the same dependent variable, it is possible to compare the factors defined by Poitras as core factors with this study’s very strongly correlated factors.

Through the analysis of Pearson’s correlation coefficient, statistically significant differences could be detected (p < 0.05) in the analyzed response patterns of our (CONMEF) instrument (Riera Adrover 2017) and the “Measurement Scale of Trust‐Building between Mediators and Disputing Parties” (Poitras 2009). As Figure Two shows, in this study it was thus possible to distinguish between the factors with a very strong significant correlation (p < 0.01), the factors with a slightly weaker significant correlation (p < 0.05), and the factors that did not display a significant correlation (Table Two).

Pearson’s correlation coefficient displayed statistically significant correlations for 80% of the factors. Due to the limitations inherent in comparing qualitative and quantitative data, multiple variables could not be included at the same time. As a result, the factors were separately correlated with the “Measurement Scale of Trust‐Building between Mediators and Disputing Parties” (Poitras 2009). This study defined the most significant factors in increasing the generation of trust, as perceived by the parties, taking values closer to 1 as indicating a higher strength.

The factors that displayed a very strong statistically significant correlation included ones associated with aspects of the questionnaire concerning goodwill and empathy, the mediator’s credibility and reputation, chemistry, and mastery. Nonetheless, aspects of the questionnaire concerning warmth and consideration, mediation principles, explaining the mediation process, and skills for the intervention showed a statistically significant correlation with the trust‐building between mediators and disputing parties variable. Aspects of the questionnaire concerning skills for building a therapeutic alliance and experience with the type of dispute displayed a nonsignificant correlation.

A comparison with Poitras’ study showed that the following factors achieved similar results in both studies (Figure Three): mastery, explaining the mediation process, warmth and consideration, and chemistry. Consequently, we can confirm Ranjay Gulati’s hypothesis (1995) that it is possible to transpose trust‐building factors to different types of mediation.

In response to the calls that have been made to test theories on trust‐building factors in mediation, this article triangulates a study conducted in Montreal (Canada) in the Balearic Islands (Spain). We have provided an outline of the main similarities between both studies. In addition, we have shown how our study extends Poitras’ model. The results of our study suggest that four factors should be taken into consideration in order to increase trust‐building between mediators and disputing parties, which will in turn foster the parties’ commitment to the mediation process. We have cited below the literature that supports the importance of these factors in order to highlight the link between the results of our study and the work of other researchers. The four factors essential to trust‐building are:

1. Goodwill and empathy: perceiving goodwill on the part of the mediator (Poitras 2009); perceiving the mediator to be motivated (Doney, Cannon, and Mullen 1998); perceiving the mediator to be patient and willing to listen to me (Goldberg and Shaw 2007); feeling listened to and understood by the mediator (Della Noce 1999; Goldberg 2005; Poitras 2009); the mediator’s commitment to the case (Deutsch 1958); and analogic communication by the mediator (looking into our eyes, leaning toward us, etc.).

2. Credibility and reputation: the mediator’s professional credibility (Doney, Cannon, and Mullen 1998; Goldberg and Shaw 2007); the mediator’s reputation (Poitras and Bowen 2002); the mediator’s appointment by a public authority and/or recognized service (Poitras and Bowen 2002); perceiving the mediator to be the right person to help overcome possible obstacles (Poitras and Bowen 2002); believing the mediator to be a fair person (Poitras, Bowen, and Byrne 2003); and the mediator’s consistency and predictability (not behaving in an unexpected way) (Doney, Cannon, and Mullen 1998).

3. Chemistry: the mediator’s manner (Deutsch 1958); the chemistry (connection) I feel with the mediator (Poitras 2009); the fact that the mediator seems to be an interesting person (attraction) (Poitras 2009); the mediator’s familiarity with the process (Poitras 2009); the creation of a relaxing setting (Poitras 2009); the mediator’s sincerity (Goldberg and Shaw 2007); and the mediator’s kindness (Lewicki and Bunker 1996).

4. Mastery: the mediator’s experience (Poitras 2009); the mediator’s capacity to understand the dispute (Poitras 2009); the fact that the mediator shows him/herself to be familiar with the dispute (Poitras 2009); the mediator’s self‐confidence (Poitras 2009); the mediator’s honesty (Doney, Cannon, and Mullen 1998); and the mediator’s capacity to manage the mediation process (Poitras 2009).

In addition to showing that goodwill and empathy, credibility and reputation, chemistry, and mastery had a significant correlation (p < 0.01) with trust‐building between mediators and disputants, our study showed a significant correlation between four other factors and trust‐building, although the correlation was not as strong (p < 0.05). However, due to the universe used in this study (n = 54) and the homogeneity of the respondents’ sociocultural backgrounds, these factors could have a greater significance in a triangulation of the test in other countries and should be analyzed when studying trust‐building. These four factors are:

1. Warmth and consideration: the mediator’s warmth (Poitras 2009); the mediator listens to me and shows consideration for my feelings and concerns (Poitras 2009); the mediator accepts that I can express my emotions (Poitras, Bowen, and Byrne 2003); the mediator understands and supports me (Suares 2002); the mediator focuses on my motivations and concerns (Della Noce 1999); and the mediator focuses on my needs (Swan, Trawick, and Silva 1985).

2. Mediation principles: having a private session with the mediator (Goldberg 2005); the mediator observes the principles of mediation (not interrupting, not suggesting, etc.) (Doney, Cannon, and Mullen 1998); the mediator’s neutrality or impartiality (Poitras 2009); the mediator does not disclose private aspects relating to the other party at private session with me, even if he/she has their permission (Suares 2002); the mediator focuses on making sure that we (both parties) understand the other person’s concerns and motivations so as to try and reach an agreement (Della Noce 1999); the mediator focuses on both parties’ shared goals (Poitras and Bowen 2002); and the mediator fosters good relations between both parties with a view to possible future negotiations (Poitras and Bowen 2002).

3. Explaining the mediation process: perceiving the mediation process to be fair (Kramer 1999); the presence of lawyers or other people in the mediation room (Deutsch 1958); the relationship with other professionals who have a high regard for mediation (Deutsch 1958); the mediator steers and supervises the mediation process (Deutsch 1958); the mediator acknowledges that he/she has made a mistake (Deutsch 1958); the mediator highlights the rules of mediation (Doney, Cannon, and Mullen 1998); and the mediator explains the mediation process in an unhurried way (Poitras 2009).

4. Skills for the intervention: perceiving the mediator’s interventions to be adequately focused (Rotter 1971); the mediator’s conflict resolution skills (Poitras and Bowen 2002); the mediator’s capacity to redefine the problem so as to highlight my interests (Poitras and Bowen 2002); clear frequent communication with the mediator (Gainey and Klaas 2005); the mediator devotes time to talking about informal matters (Poitras and Bowen 2002); and the mediator separates me from the problem (“you’re not the problem”) (Ury, Fisher, and Patton 2011).

Our results confirm the study’s initial hypothesis: The results of Jean Poitras’ study and our study can be compared in terms of the factors that increase trust‐building, even though the former involved employment mediation and our study involved family mediation; the cultural contexts were different; and our methodology was quantitive while Poitras’ methodology was qualitative.

Our results showed that the same factors that Poitras identified as fundamental in trust‐building—mastery, explaining the mediation process, warmth and consideration, and chemistry—were shown to be fundamental in our study as well. It should be noted that the mastery and chemistry factors were found to explain trust‐building to a greater extent than the factors of warmth and consideration and explaining the mediation process. However, all four factors had a statistically significant correlation with the dependent variable. This finding confirms Gulati’s premise (1995) that it is possible to transpose the factors that increase trust‐building to different types of mediation and begins to address concerns regarding the influence of cultural context on trust‐building between mediators and disputing parties.

The study’s results also demonstrate the importance of the following trust‐building factors that have been cited in five or more publications of the reviewed literature (Table Two): the mediator’s credibility and reputation, goodwill and empathy, procedural experience (skills for the intervention), and impartiality or neutrality (mediation principles). These factors show a statistically significant correlation with the dependent variable. However, the factor of experience in the type of dispute did not display a statistically significant correlation. With the exception of this factor, the results of this study confirm the importance of the factors most widely cited in the reviewed literature.

It must also be noted that the results of the study rule out the importance of the “skills for building a therapeutic alliance” factor in the generation of trust, leaving it out of the equation in trust‐building between mediators and disputing parties.

Our study reinforces Poitras’ conclusions (2009) and addresses Goldberg and Shaw’s (2007) call for more cross–cultural research into trust‐building between mediators and disputants. Our results confirm the findings of Poitras’ study and highlight interrelations in the factors that help generate a rapport: understanding, empathy, and trust (Goldberg 2005).

As with all research, this study is not without its limitations. First, it should be pointed out that the items are highly intercorrelated as a result of the single instrument. This raises the risk of a common method bias. Second, given the universe that was used in this study (the Mediation Service’s users over a period of one year), the sample was a small one. Lastly, the questionnaire we used was based on our literature review; specifically, on factors identified in prior studies as being successful in predicting the generation of trust between mediators and disputing parties. A qualitative methodology could have determined whether the bibliographical review used for this study contemplates all the factors that predict trust‐building.

Various studies have claimed that trust‐building between mediators and disputing parties is a key factor in the success of the mediation process. The main benefits of trust‐building include the stabilization of relations between the mediator and the parties and a sense of commitment to the process.

To help facilitate this commitment at the beginning of mediation processes, this study identified a series of factors that increase the generation of trust between mediators and disputing parties. They include “goodwill and empathy,” “the mediator’s credibility and reputation,” “chemistry,” “mastery,” “warmth and consideration,” “mediation principles,” “explaining the mediation process,” and “skills for the intervention.”

In contrast, when the response patterns of the two groups were compared, “skills for building a therapeutic alliance” and “experience with the type of dispute” did not display any statistically significant differences, showing a negative correlation with trust‐building in both cases. According to the results of this study, these factors have no association with the generation of trust between mediators and disputing parties.

Thus this article has contributed to further triangulations of Jean Poitras’ study in a country where mediation is starting to take off as an alternative tool in settling disputes. We identified factors that were shown to be important to trust‐building in both Poitras’ study and our study and our results show that such factors can be transposed to other types of mediation and to different cultural contexts. In finding no differences between the results of Poitras’ study and our study, we have begun to address Stephen Goldberg and Margaret Shaw’s concern about the possible influence of cultural context on the factors that increase trust‐building between mediators and disputing parties. We encourage others to undertake cross‐cultural research to explore this question further.