Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Mediation has become a common means of resolving disputes in construction for its flexible, cost-effective, and non-confrontational approach. Mediation also allows disputes to be settled voluntarily and privately without a loss of business relationships or damage to reputations (Chau 1992, 2007; Cheeks 2003; Hon 2006). Such settlements can therefore be made more rapidly and are more adaptable to the specific needs of the disputing parties in the mediation process (Cheung and Yiu 2007; Striengnitz 2006; Susskind and Ozawa 1983; Yiu and Cheung 2007). However, mediation involves a series of negotiation processes with no guarantee of success.

The actions taken by mediators are critical to mediation success. Integrity, reliability, and competence are the most important attributes of mediators that affect the level of trust that disputing parties have in them (Boulle 2001; Settle 1998). During the process of mediation, the major task of the mediator is to encourage the disputing parties, such as clients, contractors, and sub-contractors, into rethinking and modifying their positions (Kolb 1985; Madden 2001). However, the degree of influence that mediators have depends considerably on whether the disputing parties trust them (Kolb 1985). Therefore, mediators need to realise the importance of using trust-building tactics in the course of the mediation process to address any long-held and deep-seated concerns among the disputing parties (Blackstock 2001). If trust can be built, then mediators need to monitor the level of that trust (and its influence) on the disputing parties, and thus credible relationships can be established with them (Carnevale 1986; Carnevale et al. 1989). Such relationships can serve as a lubricant to avoid unnecessary hurdles in the mediation process (Boulle 2001; Torres 1991). For example, the early establishment of trust can ease subsequent inquiries into the course of mediation (Boulle 2001; Torres 1991). Time-savings can also be achieved when the disputing parties are willing to disclose confidential information or hidden agendas in the mediation process. If the disputing parties trust the mediator, then they are more likely to remain at the negotiating table, to remain committed to the mediation process, to believe in achieving successful mediation outcomes (Boulle 2001), and to behave co-operatively (Cheung and Yiu 2007). Mistrust, in contrast, discourages mediation success. The level of trust among disputing parties varies with attitude (Govier 1997). Mediators should apply trust-building tactics to generate some degree of trust in themselves and in the mediation process (Boulle 2001) to ensure that the disputing parties work toward a win-win settlement. Hence, the objectives of the study reported in this chapter were to identify the trust-building tactics used by construction mediators and to examine the efficacy of these tactics with respect to their outcomes. To achieve these objectives, three study stages were involved. In Stage I, a literature review was conducted to identify commonly used trust-building mediator tactics and mediation outcomes. Using these, a questionnaire survey was designed in Stage II to collect data from accredited mediators. A series of data analyses were then performed in Stage III to examine the efficacy of the trust-building tactics that were identified.

2 Stage I: Identification of Trust-Building Tactics and Mediation Outcomes

2.1 Trust-Building Tactics

Theoretically, the mediation process flows from one phase to the next: for example, from the mediator’s opening statement to the disputing parties’ opening statements and then to a joint session, private sessions, joint negotiations, and the final closure phase. As mediation proceeds, the mediators work to build and monitor the trust levels of the disputing parties. To do so, they employ different types of trust-building tactics in each phase of the process. Sloan’s mediation model (1998), which conceptualises the mediation process, provides an important framework for the identification of trust-building tactics in construction mediation. According to this model, which is shown in Table 19.1, the mediation process has five steps: (1) Introduction (2) Preparation (3) Issues and Trust (4) Exploration of Interests, and (5) Solutions.

Table 19.1 Brief introduction to Sloan’s mediation model (1998)
Full size table

Based on the steps in Sloan’s mediation model (1998), trust-building tactics were identified from the literature (Bercovitch and Derouen 2004; Boulle 2001; Cheung and Yiu 2007; Kolb 1985; Latz 2001; Moore 1996; Salem 2003). As can be seen from Table 19.2, a list of 18 trust-building tactics was compiled.

Table 19.2 List of trust-building tactics
Full size table

2.2 Mediation Outcomes

In this study, the outcomes refer to the results that were obtained from the trust-building tactics used in construction mediation. The generic types of outcomes, which were identified from the literature review and are listed in Table 19.3, can be reduced to: (1) trust-building between the disputing parties (2) trust-building between the disputing parties and the mediator, and (3) the negative (or positive) implications of using these trust-building tactics (Bercovitch and Derouen 2004; Boulle 2001; Butler 1991; Cheung and Yiu 2007; Kolb 1985; Latz 2001; Lui et al. 2006; Moore 1996; Salem 2003).

Table 19.3 List of mediation outcomes
Full size table

3 Stage II: Data Collection

To accomplish the research objectives, a four-part questionnaire survey was performed to collect case-specific data from construction mediators. The first part required the respondents to provide personal information, such as their sex, their age, and the number of years of experience they have had in construction mediation. Next, the respondents were asked to provide particulars about their mediated cases, such as the project nature, contract sum, and the parties involved. The last two parts of the survey were designed to identify the trust-building mediation tactics used and their respective outcomes, based on the items in Tables 19.1 and 19.2. Seven-point Likert scales were used to measure the degrees of usefulness (1: least useful; 7: most useful) and agreeableness (1: least agreeable; 7: most agreeable) on each trust-building tactic and mediation outcome, respectively. The respondents targeted in this study are accredited mediators from the General Panel of the Hong Kong International Arbitration Centre (HKIAC), which is composed of 226 accredited mediators from 30 professions. Ninety-six of these mediators who specialise in building, construction, and engineering were selected as the target respondents, and their participation was requested. If they agreed to participate in the questionnaire survey, then questionnaires were sent to them by post, fax, or e-mail. The respondents, particularly newly accredited mediators, made a number of enquiries regarding the completion of the questionnaire. Many of them wanted to know about the possibility of completing the questionnaires by referring to their experience with mediating simulated cases in lieu of reporting real mediated cases. To ensure the relevance of the responses, this suggestion was rejected. Thirty of the 96 targeted respondents completed the questionnaire survey, for a response rate of 31 %. This sample size is comparable to those in previous studies of construction mediation (Cheung and Yiu 2007; Yiu and Cheung 2007). All of the respondents are accredited mediators with a construction background, such as construction lawyers, quantity surveyors, and engineers, who actively participate in mediating construction disputes in Hong Kong. They all hold senior positions and are well-respected by the industry. For example, 36.7 % of the respondents have more than 20 years of experience in construction mediation. Fifty percent of the reported mediation cases concerned civil projects. The profiles of the respondents by experience and type of reported mediation cases are shown in Tables 19.4 and 19.5, respectively. As for the scale of the reported cases, 31.8 % of them had contract sums of less than HK$50 million. Most of the disputes (36.4 %) had arisen between the employer and the main contractor. Tables 19.6 and 19.7 show the reported mediation cases by value and the parties involved, respectively.

Table 19.4 Profile of respondents (by amount of experience)
Full size table
Table 19.5 Types of reported mediation cases
Full size table
Table 19.6 Reported mediation cases by value (in HK$)
Full size table
Table 19.7 Parties involved in reported mediation cases
Full size table

4 Stage III: Efficacy of the Trust-Building Tactics

A series of data analyses were performed on the collected data using the Statistical Package for Social Sciences (SPSS). Descriptive statistics were first performed to identify (1) the most commonly used trust-building tactics in construction mediation and (2) the most frequent outcomes achieved using them. Next, reliability assessments were performed for the Cronbach’s alpha coefficients and item-total correlations to test the internal consistency of the responses (Streiner and Norman 1997). Finally, multiple regression analysis (MRA) was employed to examine the efficacy of the trust-building tactics by relating them to the outcomes. The independent contribution of each trust-building tactic to the prediction of outcomes was investigated by the MRA regression coefficients.

4.1 Descriptive Statistics

The descriptive statistics, such as the minimum, maximum, and mean scores for each trust-building tactic and its outcomes are shown in Table 19.8.

Table 19.8 Descriptive statistics of trust-building tactics and outcomes
Full size table

In general, these descriptive statistics show that the trust-building tactics of implementing caucuses (T9), giving the disputing parties the opportunity to express themselves freely (T8), being patient and understanding the feelings of the disputing parties (T11), being well-prepared for the issues (T14), and keeping explicit promises (T18) are the most common means of trust-building. Furthermore, the mediators also reported that the improvement of relationships (O12) is the outcome most frequently achieved when using trust-building tactics.

4.2 Reliability Assessments

The internal consistency of the responses to trust-building tactics and outcomes was assessed by the Cronbach’s alpha coefficients and item-total correlations. Cronbach’s alpha provides an estimate of reliability in most situations, as the major source of measurement error is the sampling of content (Cronbach 1951). In addition, reliability that is based on internal consistency considers the sources of errors that are based on the “sampling” of the situational factors that accompany the administration of items (Nunnally and Bernstein 1994). Item-total correlation measures the relationship between an item and the total score of the set of items within the scale (Robinson et al. 1991). This correlation represents not only the relationships among the items, but also the internal consistency of the model. A low corrected item-total correlation value indicates that an item is inconsistent with the other items and is not measuring what the rest of the test is trying to measure (Ferketich 1991). The results of these two assessments are also reported in Table 19.8. As per the rule of thumb suggested by Nunnally and Bernstein (1994), a Cronbach’s alpha of 0.70 and an item-total correlation of 0.3 or above are the threshold values for these assessments. In this study, the Cronbach’s alpha coefficients for both the trust-building tactics and their outcomes exceeded the threshold value of 0.70, which indicates that the scale items are homogeneous (Bowling 1997). These rules were adopted in previous studies by Ferketich (1991), Robinson et al. (1991), and Knapp and Brown (1995). The item-total correlations for the trust-building tactics ranged between 0.19 and 0.74, and a range of between 0.20 and 0.73 was achieved for the outcomes. Two of the trust-building tactics, T3 and T15, and three of the outcomes, O2, O9, and O10, failed to meet the threshold of 0.30 (Streiner and Norman 1997) and were discarded from further analysis.

4.3 Multiple Regression Analysis

MRA, which is a statistical technique used to analyse the relationship between a single dependent variable and several independent variables (Hair et al. 1995), was employed to relate the use of the trust-building tactics to their outcomes. The efficacy of the trust-building tactics could thus be examined via the independent contribution of each tactic (i.e., the independent variable) to the prediction of the outcomes (i.e., the dependent variable). To achieve this, composite scales were calculated for each sub-group of trust-building tactics and outcomes, as shown in Tables 19.1 and 19.2. These composite scales were obtained by averaging the scores given by the respondents for each sub-group to form new sets of variables for the performance of MRA. Five composite scales of trust-building tactics were developed, each representing one of the tactics used in each step of Sloan’s mediation model (1998). Likewise, five factor scales of mediation outcomes were also developed, representing the five sub-groups of outcomes that were defined in Table 19.3. For each MRA, the dependent variable was one of the five composite scales for outcomes, and the independent variables were the composite scales of the trust-building tactics. As previously discussed, two trust-building tactics, T3 and T15, and three outcomes, O2, O9, and O10, failed to achieve the threshold values of the reliability assessments and were thus excluded from the calculation of the respective composite scales.

In light of the above, a total of five MRAs were performed. Equation 19.1 shows the MRA models.

$$ {\text{O}}_{i} = {\text{a}} + {\text{b}}_{1} {\text{T}}_{1} + {\text{b}}_{2} {\text{T}}_{2} + {\text{b}}_{3} {\text{T}}_{3} + {\text{b}}_{4} {\text{T}}_{4} + {\text{b}}_{5} {\text{T}}_{5} , $$
(19.1)

where O is the dependent variable (the composite scales of outcomes); T is the independent variable (the composite scales of trust-building tactics); and i = 1, 2, 3, 4, and 5.

The results obtained from the five MRAs are shown in Table 19.9. The R2 values for the five regression models, which represent the combined effect of the entire variant in prediction, range from 0.393 to 0.522. Comparable results were reported by Oetzel (1998), Gross and Guerrero (2000), Sharland (2001) and Cheung et al. (2006). For ease of discussion, the relative contributions of the trust-building tactics to the outcomes can be compared via the normalised regression coefficients (Cheung et al. 2006): the higher the normalised regression coefficient, the greater its contribution to the prediction of the outcome.

Table 19.9 Overall results of multiple regression analysis
Full size table

During the mediation process, the mediator works to build and maintain the trust of the disputing parties. However, the levels of trust may improve, deteriorate, or remain the same during the course of mediation (Boulle 2001). As can be seen from Table 19.9, the highest normalised regression coefficient (i.e., 0.379) of Model 3 indicates that the trust-building tactics used in Step 5 (i.e., Solutions) may not improve the level of trust among the disputing parties. Thus, the trust-building tactics used in the last step of Sloan’s model (1998) appear to have low efficacy in developing trust among disputing parties. This means that if the level of distrust is high, then the disputing parties are defensive, which makes it difficult for the mediation process to reach a joint decision. Perhaps, as supported by the MRA results obtained from Models 1 and 2, it would be more pragmatic to apply these tactics at an earlier stage of the mediation process.

Model 1 shows that the trust-building tactics used in Step 4 of the Sloan’s model (1998) (i.e., Explore Interests) are influential in developing trust among disputing parties. This can be seen from Table 19.9 in which the normalised regression coefficient of these trust-building tactics is 0.437, which is the highest among the independent variables. According to Sloan (1998), the aim of exploring interests is to help the disputing parties to identify what interests are important to them and why. The trust-building tactics used in this step, such as understanding the feelings of the disputing parties, sharing experiences, and simplifying/prioritising agendas or issues, may encourage the disputing parties to demonstrate genuine concern about the interests and enlarge the range of settlement alternatives. More importantly, the use of these tactics can encourage the disputing parties to understand each other (Blackstock 2001) and can thus facilitate communication so that the interests of each can be explored more effectively. Model 5 also suggests that the trust-building tactics used in this step can be a significant predictor of time-savings in the mediation process.

Step 3 of Sloan’s model (1998) (i.e., Issues and Trust) suggests that the mediator should encourage the disputing parties to listen to, recognise, and understand the issues of the dispute, and, most importantly, he or she should attempt to restore trust in the long run and deal with the problems of the dispute in the short run. This is thus the step that can generate trust in the mediator. The result obtained from Model 2 is consistent with these propositions: the normalised regression coefficient of the trust-building tactics used in Step 3 is the highest—0.428. This result is also supported by Boulle (2001) and Sloan (1998), who noted that such trust-building tactics as providing equal speaking time and separate meetings for the disputing parties can generate trust in the mediator, which means that these parties may be able to take risks with him or her that they would not take with each other (Boulle 2001). As mediation is a form of facilitated negotiation, the role of the mediator is critical. If the mediator can be trusted, then the disputing parties are willing to engage with him or her openly and disclose important and confidential information. This helps the mediation process to proceed effectively and eventually improves the relationships between the disputing parties. Model 4 thus supports the notion that the trust-building tactics used in this Step are also significant and contribute to the improvement of the relationship between the disputing parties.

5 Chapter Summary

The appropriate use of trust-building tactics can have an immense impact on mediation outcomes. The aim of the study reported in this chapter was to identify the trust-building tactics used in construction mediation and to examine the efficacy of these tactics with respect to their outcomes. The research design was based on the mediation model of Sloan (1998), and, with the use of MRA, the key findings of this chapter can be concluded as follows. (1) The trust-building tactics used in Step 4 (i.e., Explore Interests) of Sloan’s mediation model (1998) are influential in developing trust among disputing parties and can also act as a time-saving tool in the mediation process. (2) Mediators can earn trust by adopting the trust-building tactics used in Step 3 (i.e., Issues and Trust) of the Sloan’s model (1998) and can also improve the relationship between the disputing parties. (3) The trust-building tactics used in the last step (i.e., Solutions) of Sloan’s model (1998) appear to have low efficacy in developing trust among disputing parties.