No description available.

In a review on conflict strategies, Putnam and Poole (1987) conclude that relationship variables such as organisational position and interdependency account for the most variance in the choice of conflict strategies. Despite these findings, however, there is a general lack of attention to relationship outcomes in conflict management by researchers (Knapp, Putnam, and Davis, 1988). This reflects the strong influence of economic theory in conflict studies as well as a Western perspective in which individual outcomes are emphasised above interpersonal relations (Wall & Callister, 1995). To address this problem, in this thesis I chose to assess the effect of Past-Relationship, Other's-Strategy, Hierarchy, and Stake on Chinese manager's behaviour in simulated conflict over welfare issues. The thesis consists of the following seven chapters. Chapter One, Introduction, identifies the need to address relationship considerations in conflict management, and the PRC Chinese were chosen as subjects for this research because of their emphasis on interpersonal relations. Past-Relationship, Other's-Strategy, and Hierarchy were chosen to represent the ongoing interpersonal relationship in the past, present and future; and Stake was used to represent substantive considerations in conflict. These factors were assessed for their effects on people's use of five conflict modes: Dominating (DO), Compromising (CO), Obliging (OB), Integrating (IN), and Avoiding (AV). Although these five conflict modes are based on a model developed in the West, they are applicable to the Chinese as well despite the recent critics by some authors. The limited research on Chinese interpersonal conflict has over-relied on self-report questionnaires and lacks the context and substance of conflict. To avoid these pitfalls, the current research took an experimental approach to solicit subject response to simulated conflict of interests. Chapter Two, Hypotheses and research design. Five sets of hypotheses were developed. Stake, Past-Relationship, and Other's-Strategy were expected to affect the use of conflict modes in their own respective ways; Hierarchy's effects were to be moderated by Stake; and an interaction between Other's-Strategy, Past-Relationship, and Hierarchy was also expected. To test these hypotheses, three sources of data were collected: behaviour in conflict simulation, behaviour rationales, and cases of real-world conflict. The conflict issues in the simulation were limited to the allocation of welfare benefits. The range of options and potential outcomes were made explicit to the subjects. Hierarchy, Past-Relationship, and Stake were operationalised through 12 conflict scenarios, which were then combined with Other's-Strategy, giving rise to a total of 48 experimental conditions. A total of 384 Chinese managers participated in the conflict simulation. Chapter Three, Conflict Mode Simulation (CMS). This chapter introduces the development of the experimental instrument, the CMS. Based on the analysis of Mode Game (Cosier & Ruble, 1981), the CMS was created as an experimental instrument for research of the five conflict modes in general and for the hypothesis testing in this study in particular. It allows participants five conflict modes to deal with conflict scenarios; the payoffs of different combinations of modes were logically derived; the 3-2-2 settlement structure enables the CMS to accommodate complex behaviour patterns; and the combination of unitary rules with multiple conflict scenarios controls confounding effects, and gives CMS the flexibility for different research topics. Chapter Four, Data collection and analysis of the simulation data, discusses hypothesis testing through the conflict simulation. The data fit the theoretical model satisfactorily in general. The five sets of hypotheses were largely supported by the simulation data. 1) As the Stake increased, the use of DO and IN increased and the use of OB decreased. The use of AV peaked on Medium-Stake issues. 2) Hierarchy strongly affected the use of DO and OB on High-Stake issues, but that effect diminished as Stake decreased and no difference between hierarchy groups was found at the Low-Stake level. 3) The Indebted group used more OB and fewer DO than the Wronged group, but the uses of IN and AV were not related to Past-Relationship. 4) Other's-Strategy affect behaviour through reciprocating, suppressing, promoting, and learning effects. 5) The same mode (Compromsing) used by different people under different situation had different results. It was also shown in post hoc analyses that relationship concerns played important roles even when faced with the rivalry of substantive concerns. Moreover, the negative effect of combative behaviour was stronger than the positive effect of good deed. Several other important behavioural patterns were also discussed. Chapter Five, Analysis of the rationales of CMS behaviour. Twenty-six major rationale themes were reported by subjects to account for their behaviour in CMS. The report of competitive vs. conciliatory themes depended mainly on Past-Relationship and Stake. The Indebted group reported more conciliatory themes and fewer competitive themes than the Wronged group; and the High-Stake group reported more competitive themes and fewer conciliatory themes than the Med/Low-Stake group. Hierarchy's effect on subject rationale was weak. It was also discovered that subject sensitivity towards Past-Relationship shifted according to Stake. The four modes of CO, OB, IN, and AV shared the same pool of conciliatory themes, which distinguished them from DO, but there were important differences among them as well. Chapter Six, Analysis of self-reported cases of conflict. Fifty-two cases of real world conflict comparable to the CMS conflict scenarios were collected. Analysis shows that Past-Relationship, Stake, and Other's-Strategy affected behaviour in these conflict cases in ways in-line with their respective hypotheses. Hierarchy, however, did not show reliable effects on subject behaviour. Subject description on different types of competing enables the in-depth analysis of the DO mode. The nonsymmetrical effects of Past-Relationship, Stake, and Other's-Strategy manifested in CMS are also observed in these real conflict cases. Chapter Seven, Discussion and conclusion. After an overview of the results of hypothesis testing in different data sources, the implications of some particular issues are discussed. 1) Although the two-dimension model fitted the CMS data well when a variety of different conflict situations were assessed together, in particular situations a hierarchical model is more accurate. 2) Evidence from different data sources confirms that relationships in the past, present and future all have a role to play in the conflict of interests. 3) In terms of the relative importance of different factors, Past-Relationship and Other's-Strategy are at least as important as Stake, whereas Hierarchy is the least important in the given situation of this study. 4) The weak effect of Hierarchy reflects the constraint by status ethics (Hwang, 1991), the interdependence between superior and subordinates, and the broader power base of the highly educated subordinates. 5) Although avoidance was perceived positively, it was not used much when specific conflict issues were given. 6) The Conflict Mode Simulation (CMS) has proved to be indispensable for many of the findings in this study, and it has great potential as an experimental tool for research of conflict modes in general. Finally the limitations of this study and the direction of future research are discussed.