Supporting Boundary Spanning Roles in Water Boards: Finding a New Balance to Accommodate Change

New Balance to Accommodate Change

Master thesis, MscBA, specialization Change Management

University of Groningen, Faculty of Management and Organization

CASPER ELLEN

Student number: 1256890

Koninginnelaan 31

9901 HL Appingedam

tel.: 06-11001680

e-mail: casellen@hotmail.com

Date of submitting: December 13

, 2006

1

Supervisor

Prof. Dr. H. van Ees

2

Supervisor

Dr. B.J.M. Emans

Supporting Boundary Spanning Roles in Water Boards: Finding a

New Balance to Accommodate Change

ABSTRACT

INDEX

1) INTRODUCTION 3

2) THEORETICAL BACKGROUND 5

2.1 Role theory, ambiguity and role conflict 6

2.2 Boundary spanning activities in role theory 7

2.3 Difference between managerial and non-managerial boundary spanners 8

2.4 Present study in relation to previous work 9

3) CONCEPTUAL FRAMEWORK 10 3.1 Organizational factors 10 3.1.1 Formalization 10 3.1.2 Autonomy 11 3.1.3 Organizational fit 12 3.2 Personal factors 13 3.3 Interpersonal relations 15

3.3.1 Interpersonal relations in an internal and external context 16

3.3.2 Power, dependence & trust 17

3.3.3 Communication & feedback 18

3.4.4 Participation 19 4) RESEARCH METHODS 20 4.1 Data collection 20 4.2 Measures 22 5) RESULTS 23 5.1 Organizational factors 25 5.1.1 Formalization 25 5.1.2 Autonomy 26 5.1.3 Organizational fit 27 5.2 Personal factors 29 5.2.1 Competencies 29

5.2.2 Personality & Attitude 31

5.3 Interpersonal factors 31

5.3.1 Interpersonal relations in an internal and external context 31

5.3.2 Power, dependence & trust 32

5.3.3 Communication & feedback 34

1) INTRODUCTION

The Dutch water management landscape is changing. Nowadays, key complex problems in water management more and more need to be tackled through collaborative intervention. Climate changes, urbanization, European water quality standards, together with pressure on water boards to operate more efficiently, require policy changes. Although differences exist between the various tasks water boards have in inter-organizational activities, the common element is that water boards1 have become more dependent on other organizations to realise their own goals. This change in relations requires a coherent change plan for the water boards to change themselves as well. In February 2001, the State, Inter-provincial deliberation, the Union of Water Boards and Council of Dutch Municipalities signed the Starting Agreement Water Policy 21st century. Two years later the results of the cooperation are incorporated in the National Governing Agreement. Water boards state that they are working hard to become the water board of the 21st century.

The committee that developed the Starting Agreement Water Policy 21st century advised that governmental and societal organisations have to work more together and try to reach agreement on a voluntary basis to achieve policy goals. Next to monitoring and control, water boards nowadays also need to advise to, cooperate with, and convince partners about the importance of good water management and shared goals. Therefore, administrators nowadays more and more communicate with other organisations as an integral part of their job, and thereby “span” boundaries. I refer to these administrators as “boundary spanners”. Changes in organizational roles may result in role conflict and ambiguity when there is a lack of accommodating change. House and Rizzo (1972) argue that ambiguous or frequently changing directives, a lack of clear policy guidelines, performance standards and feedback, may result in role ambiguity and conflict. Prior research in the project “WIDS” (Water in the city), has shown that inter-organizational cooperative efforts in water management to build bridges between organizational goals are not always successful, because perceptions of roles between individuals and/or organizations can be rather different (Van Ees, Six, 2006).

Cooperating to meet (inter-)organizational goals requires water boards to take into account multiple stakeholders. Usually, innovation lies in the improvement of a product, technology or service (Daft, 2000). In the water management sector, there has particularly been innovation in the process of inter-organizational cooperation in order to reach policy

1

goals more effective- and efficiently. Decision-making has shifted towards consensus formation and trust building has become more important. Administrators in water boards are increasingly aware of their advisory role in inter-organizational activities, while they remain responsible for monitoring and controlling of a particular area. The administrators that perform these daily activities often cooperate with outsiders and coordinate internally. While prior literature on boundary spanners has had its focus on strategic management and strategic agents in the context of strategic alliances, boundary spanning activities are nowadays often performed by non-managerial employees as part of their day-to-day operations. These boundary activities require skills, attitude and behaviour different from internal roles. Furthermore, organizations can accommodate change to support administrators performing boundary spanning activities by reducing the potential for conflict and ambiguity.

Much of the research on inter-organizational relations is on motives, contingencies and structures, and less on processes (Williams, 2002). The purpose of this study is to learn more about the organizational, interpersonal and personal change processes that accommodate inter-organizational cooperation. A new approach to study this change process towards more pro-active cooperation is to apply role theory (Kahn et al., 1964) to identify factors non-managerial administrators in boundary spanning positions experience as inhibiting optimal inter-organizational performance. The research question is: How can the role episode model

of Kahn et al. (1964) be applied to illustrate the perception of boundary spanners on how antecedent factors of role conflict and ambiguity influence their performance in inter-organizational activities?

Once these factors are identified, I can answer the following remedial question. What

changes can be made on an organizational, personal, and interpersonal level to change water boards from a traditionally inward focused organization towards a 21st century water board?

The results of this study can be applied to provide advice on how to further improve the performance of administrators in boundary spanning roles.

2) THEORETICAL BACKGROUND

2.1 Role theory, role ambiguity and role conflict

The theoretical background of this research stems from role theory and literature on boundary spanners. In this context, a “role” can be defined as a set of expectations applied to the incumbent of a particular position by the incumbent and by role senders within and beyond an organization’s boundaries (Banton, 1965). Kahn et al. (1964) defined the “role set” as all persons whose expectations are relevant to the performance of the focal role. Role senders influence the focal person by regulating their behaviour in accordance with the role expectations they hold.

The basic role episode model (Kahn et al., 1964) shows the process of role sending and receiving (see Figure 1). Organizational, personal and interpersonal factors influence this process. The expectations of others with respect to the incumbents behaviour is referred to as the sent-roles (I). Role pressures are exerted to make the expectations and perceived behaviour congruent (II). The total pattern of sent role pressures yields the potential for conflict, while the total of objective ambiguity is derived from the available information relevant to the position. The sent-roles are then interpreted by the incumbents receive role, which will determine the amount of conflict and ambiguity experienced (III). The final step in this process is the response of the focal person towards his role senders (IV).

---Insert Figure 1 about here---

personality (8) and interpersonal relation (9) with role senders can change over time or abruptly, e.g. a heated discussion may change interpersonal relations permanently.

In role theory, role ambiguity and conflict are two important constructs that are found to inhibit role performance (Kahn et al. 1964; Rizzo et al., 1970; Miles, 1976; Senatra, 1980). Role ambiguity is a direct function of the discrepancy between the information available to the person and that is required for adequate performance of his role (Kahn, 1964). Rizzo et al. (1972) found that more goal consensus and clarity led to less ambiguity. More recently, Bray and Brawley (2002) found individuals who reportedhigher role clarity also reported higher role efficacy and performedbetter than those with lower role clarity. Two explanations have been offered for why role ambiguity has detrimental effects. Kahn et al. (1964) argued that although there are individual differences in terms of a person's ability to tolerate ambiguity, all employees find the state of being uncertain to be stressful. King & King (1990) suggested that the negative consequences of role ambiguity result from the fact that role ambiguity interferes with goal accomplishment. Role stress can be seen as the result of suboptimal role performance (King & King, 1990; Sawyer, 1992). There are five widely accepted dimensions of role ambiguity, goal, responsibility, process, priority and behaviour ambiguity. Process ambiguity is about how to achieve organizational objectives, while behaviour ambiguity refers the clarity about which behaviour will lead to the needed or desired outcomes. Various forms of ambiguity have been given different names by various authors (e.g. Breaugh & Colihan 1994). Performance criteria ambiguity is a form of goal/ expectation/ performance ambiguity, while method ambiguity is a mix of behaviour and process ambiguity.

2.2 Boundary spanning activities in role theory

The organizational structure defines the position of the role senders and focal persons. A formal organization is system of positions and their associated roles (Rogers & Molnar, 1976). A boundary position is one for which some role senders are located in a different system from the focal person (Kahn et al, 1964). A boundary spanner is defined as someone who occupies a boundary position at the periphery of an organization. It is important to study the unique nature of boundary spanning employees’ work environments and tasks, given that their jobs and role senders are often quite different from those investigated in prior research studies (e.g. the experiences of salespersons and manufacturing workers). Kahn et al. (1964) distinguish between two dimensions of boundary relevance. These are the amount of time spend on role-related contact with persons outside his or her work unit and the importance of such contacts for effective performance of the focal person. Evidence has been found that the greater the extent to which people engage in boundary spanning activities the greater the experienced role conflict (Kahn et al., 1964; Miles, 1976, Miles, 1977). 68 Per cent of the focal persons in boundary spanning position are above the median index of objective role conflict, while the corresponding figure for positions away from the boundary is only 32 per cent (Kahn et al. 1964). Note though that they used a crude measure of boundary relevance by treating boundaries within and between organizations, as well as nearby or distant, the same.

can be resolved internally, because an organization is only likely to formulate a policy on how to resolve conflicts with some outside agency when such conflicts are extensive and serious (Kahn et al., 1964).

Miles (1977) findings also suggest that the nature and intensity of role conflict experienced are a function of the role a person occupies in an organization. For boundary-spanners, an increase in role-set distance is less stressful in terms of role ambiguity than for internals regarding both expectations and evaluations (Miles, 1977). Organizational distance is defined as the number of distinct intra- or inter-organizational boundaries which separate focal person and role sender. Miles (1977) also found that while boundary spanners clearly have the highest levels of role conflict, they appear to benefit from a threshold effect. Kahn et al. (1964) argued that boundaries between departments reduce the role sender’s knowledge of the focal job, and that the role sender’s ignorance of the focal job was directly related to the degree of role conflict and ambiguity experienced by the focal person.

Contacts with outsiders do not necessarily increase role conflict and role ambiguity; an exchange of information may improve mutual understanding. Furthermore, experiences with similar conflicts improve the boundary spanners coping technique (Kahn et al., 1964). However, when contacts include cooperation with dissimilar focus or interests, conflict may arise over what is a fair exchange, or over demands that one organization makes on another (Rogers, 1974). Singh & Rhoads (1991) argue that Rizzo et al.’s measure (1970) of role ambiguity often used in prior research on role conflict and ambiguity (Jackson and Schuler, 1985) under identifies the role ambiguities faced by boundary spanners. Specifically, managers in other departments, co-workers, and ethical conduct on the job (e.g. in interacting with organizations and customers) are additional facets Singh & Rhoads (1991) found.

2.3 Difference between managerial and non-managerial boundary spanners

managerial boundary spanning or cross-cutting posts and administrators in mainstream jobs whose boundary spanning activities are becoming more and more an integral part of their jobs. The applicability of managerial boundary spanning literature on non-managerial boundary spanners could be a topic of debate. I expect that the two types of boundary spanners largely require the same traits, skills and personality to be able to fulfil their boundary spanning activities satisfactory. However, it remains to be seen how much power and authority non-managerial administrators performing boundary spanning activities have. Therefore, prior boundary spanning literature is expected to be only partially applicable.

2.4 Present study in relation to previous work

I expect the role episode model of Kahn et al. (cf. Kahn et al., 1964) to be unable to fully grasp the complexity of inter-organizational activities, because it treats internal role senders the same as eternal role senders. I share the criticism of Miles (1977) that Kahn et al. (1964) routinely selected role senders on the basis of a priori judgment, and thereby neglected to include outsiders. External role senders should be included in the role set when their expectations are relevant to the performance of the administrator in boundary spanning activities, even though may not have formal authority over these administrators. It is therefore not surprising that Kahn et al. (1964) reported that 88% of all inter-sender role conflicts came from above, when outsiders, from which half of the conflicts were reported, were excluded. The processes of intra-organizational and inter-organizational role sending and receiving are expected to be different. Therefore, the role episode model of Kahn et al. (1964) may need to be altered to include external role senders explicitly when these are found to be sufficiently different from their internal counterparts. The present study primarily extents prior literature in role theory by re-evaluating organizational, personal and interpersonal factors involved in role conflict and ambiguity in a water management boundary spanning context. The practical goal is to advise structural interventions to reduce role conflict and ambiguity of administrators performing boundary spanning activities in water boards.

An integration of prior literature suggests a set of organizational, personal and interpersonal factors that should be studied in a boundary spanning context, which are summarized in Figure 2. The distinction between three sets of factors is derived from the model Kahn el al. (1964) as shown in (see Figure 1).

3) CONCEPTUAL FRAMEWORK

3.1 Organizational factors

Organizational factors have been extensively studied, and of often found, as a source of role conflict and ambiguity (House & Rizzo, 1972; Rogers & Molnar, 1976; Morris, Steers & Koch, 1978; Sawyer, 1992). Two organizational factors have been extensively studied in this context. These are formalization (House & Rizzo, 1972; Rogers & Molnar, 1976; Morris, Steers & Koch, 1978) and autonomy (Rogers & Molnar, 1976; Sawyer, 1992). I also include organizational fit, which the entails coherence between organizational goals, leadership, and culture.

3.1.1 Formalization

A frequently researched antecedent of role conflict and role ambiguity is formalization, which can be seen as the degree to which standard practices, policies, and position responsibilities are formalized explicitly (House & Rizzo, 1972). The formalization research stems from the organic-mechanistic perspective. Burns and Stalker (1961) and Lawrence and Lorsch (1967) contrast organizational characteristics in terms of an organic-mechanistic typology. They argued that organizations in stable environments will be more effective if it is structured in a mechanistic, rather than organic matter. Organic organizations are said to deemphasize hierarchical structure, have a high degree of trust openness and implicit rather than explicit goals, and rely on unprogrammed problem solving (House & Rizzo, 1972). Since an organic and a mechanistic organization are two ends of the same continuum, the mechanistic organization can be characterized in “opposite” terms, e.g. tight hierarchical structure, little autonomy and explicit directions, low mutual trust, win-lose bargaining structures, and a primary orientation towards the organization (House & Rizzo, 1972). According to various researchers (House & Rizzo, 1972; Rogers & Molnar, 1976; Morris, Steers & Koch, 1978), formalization correlates to less role conflict and ambiguity. Jackson and Schuler (1985) find the correlation between formalization and role ambiguity much higher (-.49) than with role conflict (-0.11).

intra-role and interpersonal conflicts in conservative, rules-oriented, bureaucratically oriented organizations with emphasis on formalization and standardization. Intra-role conflicts arise because his organizational position demands him to proceed along administrative paths governed by existing rules, while at the same time he is expected to be oriented towards innovative decisions. Therefore, I expect that boundary spanning activities, as opposed to internal roles, cannot be supported through formalization. Such innovative roles are nonetheless functional for the organization as a whole, because it provides adaptive flexibility to a bureaucratic structure (Kahn et al., 1964).

I expect a more organic, rather than mechanistic structure, to support boundary spanning activities. However, attempts toward more organic practices may, although necessary, result in more ambiguity and conflict, especially in the short run. Therefore, other supportive practices to reduce ambiguity are expected to potentially improve performance by acting as a substitute for formalization. Stamper and Jolke (2003) hypothesized and found that organizations that care about employee well-being are probably more likely to provide organizational support by specifying and clarifying job expectations and norms in order to better prepare them for work assignments, thus directly reducing the amount of conflict and ambiguity associated with various tasks.

3.1.2 Autonomy

Autonomy here refers to an administrator’s authority to make decisions concerning the means to achieve organizational goals. Sawyer (1992) found autonomy to be antecedent to process clarity. In their meta-analysis, Jackson and Schuler (1985) found autonomy to be negatively correlated with role ambiguity and uncorrelated with role conflict. However, they found considerable variance across studies that cannot be accounted for by statistical artefacts. An explanation for this variance might have been provided by Miles (1977). While researching the effect of organizational distance of role senders on role conflict, Miles (1977) found sharp differences in strength and relations between the boundary spanning and internal roles. He argues that for role sets with large distance between the role senders and boundary spanner, e.g. due to boundaries between divisions and/or organizations, may provide the latter with the autonomy to largely define his own role expectations.

authority was also associated with higher levels of role conflict. Singh (1993) found that greater autonomy for boundary spanners invariable results in lower perceptions of role ambiguity.

Williams (2002) found that a poor partner is perceived to be one who often has to ‘report back’ to the home organization. ‘The more effective partners are those who are empowered, within certain parameters, to engage constructively with other partners’ (Williams, 2002). Autonomy refers here to an administrator’s authority to make financial and strategic decisions that set goals and the way for achieving them. Too low autonomy can lead to strongly conflicting demands between the home organization and the external role sender, while too high autonomy may cause ambiguity, because the boundary spanner could become unaware of how the organization would prefer him/her to act, behave, and represent the organization.

Based on the notion that water boards are traditionally hierarchical organizations, I expect autonomy to be too low rather than too high.

3.1.3 Organizational fit

Organizational fit is defined as the coherence between organizational goals, leadership and culture. Organizational culture can be defined as the particular set of values, beliefs, customs and systems that are unique to an organization (Burnes, 2004). Singh, Verbeke, and Rhoads (1996) studied the influence of organizational practices on role stress for marketing-oriented boundary spanners. They argue that from a boundary spanner’s perspective, organizational practices can be viewed as favourable or unfavourable. They used five continuums of organizational practices, closed-open, parochial-professional, job-employee, results-process, and internal-external focus, of which the latter of each time represents the inherently favourable dimension expected by boundary spanners. However, they argue their findings also suggest that intervention programs to purge jobs role ambiguity and conflict may find jobs that are too limiting have been created. The existence of a simple and sovereign relationship across various contexts is therefore ill-conceived. I expect that the changing nature of administrators’ activities in water boards towards more cooperation requires a new balance between the continuums to further support administrators performing boundary spanning activities.

A role change can cause role ambiguity and conflict when it is not evenly supported by all members of the organization. New behaviours, works methods, policies, visions etc. have to be tried and evaluated. Cameron and Quinn (1999) developed an organizational culture framework using a theoretical model of “Competing Values” (see Figure 4). This framework refers to a matrix looking at whether an organization has a predominant internal or external focus and whether it strives for flexibility and individuality or stability and control. Hence there are four dominant culture types: clan, adhocracy, market, and hierarchy. The leadership type underpinning each of these cultural types provides an understanding of the expected behaviour in each cultural type (see Figure 5). Each quadrant represents basic assumptions, orientations, and values. Notice the word dominant, which entails that the other cultures are often to a lesser extent present as well. Yukl (1989) uses “modern” and “post-modern” to distinguish between an “old” and “new” leadership style. I believe that the leadership style must be coherent with the organizational culture and vision towards the future of the organization in order to present the same message to boundary spanners. For example, if the organizational goal is innovation, evaluation based on efficiency is likely to present conflicting or ambiguous information as to what is considered “high performance”.

---Insert Figure 5 about here---

Inter-organizational capacity is unlikely to flourish in organizational structures that are based on hierarchical control and power (Williams, 2002). On an organizational level, given that problems have become cross boundary in nature, I expect a post-modern, rather than a classical form of organization more suitable to achieve current and future goals. This entails a shift in organizational action from preoccupation with intra-organizational issues to inter-organizational capacity building, and a shift from theory X to Y leadership (McGregor, 1960).

3.2 Personal factors

Pearce (1981) argues that individuals in an unusual setting, those for whom job-related expectations are changing, and when an organizational setting is changing, established behaviour-outcome expectancies are questioned. Old behaviour does not provide the expected consequences, and new behaviour has to be tried and evaluated. Furthermore, persons in creative roles are frequently expected to possess ancillary skills in human relations.

Webb (1991) refers to boundary spanners as ‘individuals who are especially sensitive to and skilled in bridging interests, professions and organizations’. Trevillion (1991) views boundary spanners as ‘cultural brokers’ who need to understand another’s organization and to make a real effect to empathize with, and respect another’s values and perspectives. Managing expectations has become more important over recent years in water management, because water boards, and their representatives, cannot make one sided demands anymore. Complex public policy problems require new ideas, creativity and an unlearning of professional and organizational conventions and norms (Williams, 2002).

Next to managing expectations, managing trust and building sustainable relations are essential role requirements to an effective boundary spanning working relationship. A complex mix of competencies, personality and attitude is required to visualize reality from the perspective others, and all the qualities, skills, traits, and experience that are required to cooperate effectively. Effective boundary spanners are able to cultivate inter-personal relationships, communicate clearly and effectively, and appreciate the interdependencies in their job (Williams, 2002). An atmosphere of inter-organizational working needs to be set within decision making models that are premised on consensus, equality and win-win solutions. In prior boundary spanning literature, the focus has been on managerial boundary spanners. Non-managerial boundary spanners are expected to be similar to managerial boundary spanners in the sense that they both have to bring partners together, break through red tape and see things in a different way. Therefore, important competencies administrators in boundary spanning activities require include managing expectations, managing trust and building sustainable relations.

Locus of control reflects an individual’s perception of causality. “Internals” believe that events result primarily from their own behaviour and actions. They are more likely to hold themselves responsible when faced with ambiguity and conflict. In contrast, “externals” believe that events outside their control such as powerful others, fate, or chance primarily determine events. Externals are more likely to blame others for his difficulties when responding to the same situation. An external locus of control is correlated with high ambiguity and conflict scores (Organ & Greene, 1974; Behrman et al., 1981, Jackson & Schuler, 1985). Internals tend to be better informed about their jobs than externals. Furthermore, internals rely more upon their own set of definitions and are more active in seeking information and knowledge concerning their situation, and are therefore less prone to unclear role senders. As a result, internals are more adaptive (see Figure 1, arrow 5) and respond with much less defensive and more task-oriented coping than for those who have a high external locus of control (Anderson, 1997). A focal person’s personality itself may change over time as well (see Figure 1, arrow 8). Since boundary spanners are expected to be confronted with more ambiguity and conflict (Kahn et al. 1964; White, 1974; Rogers & Molnar, 1976; Miles, 1976; Miles, 1977; Pearce, 1981), it is expected an internal locus of control will facilitate coping.

3.3 Interpersonal relations

The interpretation of pressures exerted by role senders upon the focal persons depends on the interpersonal relations between them (See Figure 1, arrow 7) Changes in the focal person’s respect, trust and liking towards his role senders create significant changes in the interpersonal relations (See Figure 1, arrow 9). Such a change in interpersonal relations will in turn affect the pressures role senders exert in future role sending behaviour (See Figure 1, arrow 6). The term interpersonal relations refer to the patterns of interaction between a person and his role senders. These patterns of relationships can be categorized along several dimensions; some of these interactions might be stemming from the formal structure of the organization, while others stem from the informal interaction. The focus of prior literature has always been on internal role senders. However, I expect that this is not correct, because external role senders influence focal persons through the concept of functional dependence (Kahn et al, 1964). Therefore, as shown in section 3.3.1, the inter-organizational structure should also be taken into account.

trust in the cooperativeness and benevolence of the other, and attraction or liking; (3) dependence on the other; and (4) the style of communication between the focal person and his role senders. The first three dimensions are strongly interdependent and will therefore be presented simultaneously in section 3.3.2.

3.3.1 Interpersonal relations in an internal and external context

White (1974) agues that administrators involved in inter-organizational relations characteristically lack formal authority over those outside their organization. Therefore, role senders who have formal authority are likely to present their demands differently from others. Power is defined as the ability to influence others to believe or behave, or act. While only a small set of persons that have interest can put significant pressures on the focal role by using formal power, the population or role senders for a given focal person include all persons whose expectations are relevant to the performance of the focal role. A person’s behaviour is not a matter of his supervisor alone. Many others have a vested interest in what he does and how he does it. The concept of functional dependence (Kahn et al., 1964) reflects that a person’s ability to accomplish tasks adequately depends on others.

Although formal power is highly concentrated in the hand of direct superiors, the functional dependence is distributed throughout multiple role senders. Municipalities are expected to be able to influence boundary spanning administrators in water boards through functional “performance dependence”, since they partly determine the effective performance of boundary spanning administrators of water boards. Role sending and receiving processes that exist through this performance dependence are expected to be different from ones based on the internal formal hierarchy, because interpersonal and inter-organizational relations are expected to act as a substitute when there is no clear formal hierarchical structure between organizations.

Molnar (1976) found that inter-organizational variables accounted for the largest amount of variance in role conflict, while intraorganizational variables accounted for the largest variance in role ambiguity. Apparently, inter-organizational relations influence role ambiguity and conflict of boundary spanners. Therefore, it seems reasonable to assume that inter-organizational relations influence role processes as well. As a result, I expect that the role episode model of Kahn et al. (1964), as presented in Figure 1, may need to be adapted when used in a complex collaborative context to explicitly incorporate external role senders apart from internal role senders.

3.3.2 Power, dependence & trust

A common typology of power is the five basic power types of French and Raven (1959). These are expert power, legitimate power, referent power, reward power and coercive power. Expert power is based on what one knows; legitimate power results from being elected or appointed a position of authority; referent power stems from affiliations we make to entities we belong to; reward power is based on the right to offer or deny rewards; coercive power exists when the threat or use economic or political force. Coercion typically induces distrust. Boundary spanners need to appreciate the interdependencies in their job and ‘unlearn’ old coping behavior (Williams, 2002). Cooperation between water boards and municipalities, who have limited authority over each other, have different work methods and interests, and relatively little rewards to offer, are complex and fragile (Van Ees & Six, 2006). Individual boundary spanning representatives of organizations that have grown more dependent on each other are expected to rely more on personal sources of power and influence rather than formal authority (Rogers & Molnar, 1976). Lacking formal power over outsiders, a boundary spanning administrator is expected to rely heavily on respect, trust and liking that can be generated among outsiders. In other words, an administrator is expected to rely on personal sources of power and influence rather than formal authority in boundary spanning activities.

power as a control instrument. In other words, when there is more trust, there will be less need for control (Das en Teng, 1998). Hardy, Phillips & Lawrence (1998) distinguish between real and simulated trust, and suggest that power can be hidden behind the rhetoric of “collaboration” to promote vested interests.

3.3.3 Communication & feedback

Communication processes have been shown to correlate with both role ambiguity and role conflict (Schuler, 1979). Kahn et al. (1964) do not distinguish among dimensions of communication between the focal persons and the role senders. Only the continuum of frequency is incorporated in their model. They found a moderate but highly significant correlation suggesting that more communication clarified role expectations and reduced role conflict while less communication had opposite effects (Kahn et al. 1964).

Kahn et al. (1964) argue that there can be a vicious cycle between communication and role conflict and/or ambiguity. A focal person that experiences low role conflict and/or ambiguity is likely to augment his relationship with the role senders, because the focal person’s levels of trust, liking, and/or respect for the role senders increase. On the other hand, high role conflict and/or ambiguity perceived by the focal person may cause the focal person to withdraw from the relationship with role senders. The withdrawal lowers the frequency of communication between the focal person and the role sender, resulting in even higher levels of role conflict and ambiguity.

Reactions may produce changes on future role sending behaviour directly (Figure 1, arrow 2), but also through personality factors (Figure 1, arrows 8 and 4) and interpersonal relations (Figure 1, arrows 9 and 6). Sawyer (1992) found that task feedback and agent feedback were antecedent to process clarity. A focal person can attempt to initiate communication with his role senders, increasing the effectiveness of communication with them and the likelihood of joint problem solving. Such direct feedback may improve coping behaviour or change the demands made upon him. Feedback cycles may also be indirect (e.g. periodical assessments).

understandable. In terms of feedback, I expect that boundary spanners that receive feedback from outsiders will be more aware of their perception and can interpret this information to improve his performance as a boundary spanner.

3.4.4 Participation

Participation in decision (PDM) is a relatively weak form influence compared to autonomy. There is evidence that that PDM is negatively causally related to role ambiguity and conflict (Schuler 1977; Jackson, 1983; Morris, Steers and Koch, 1979). Jackson (1983) hypothesized that PDM determines two aspects of a person work situation: influence over others’ role expectations, which could be used to reduce role conflict, and availability of information about role expectations, which could reduce role ambiguity. Morris, Steers and Koch (1979) also found that it appears that the key to reducing role ambiguity and conflict may lie in providing employees a larger voice in decisions affecting their jobs. Chonko (1982) argues that a more participative supervisory style can help to reduce role ambiguity, because it improves communication. The benefits of participation include clarity and more applicable written rules and procedures concerning the role to be performed. Similar to problems that may exist due to low autonomy, PDM requires boundary spanners to often ‘report back’ (Williams, 2002) to the home organization. Therefore, I expect that even much PDM is an inadequate substitute for autonomy in boundary spanning activities. To summarize, I provide a short of all expectations mentioned in Table 1.

TABLE 1

Summary of expectations derived from the literature review

§

§ 2.2 Boundary spanners are expected to be confronted with more role ambiguity and conflict than administrators without boundary spanning tasks.

§ 2.3 Non-managerial boundary spanners are expected to be similar to managerial boundary spanners and largely require the same traits, skills and personality.

§ 3.1.1 I expect that the effective performance of boundary spanners cannot be supported through formalization.

§ 3.1.1 Supportive practices to reduce ambiguity are expected to potentially improve performance by acting as a substitute for formalization.

§ 3.1.2 _{I expect the autonomy of boundary spanning administrators to be too low.}

§ 3.1.3 I expect that the changing nature of administrators’ activities in water boards requires a new balance _{to achieve organizational fit and thereby support administrators performing boundary spanning} activities.

§ 3.1.3 I expect a post-modern, rather than a classical form of organization is more likely to be in tune with the policy challenge in water management.

§ 3.2

and competencies from administrators in non-boundary spanning roles. § 3.2 _{It is expected an internal locus of control will facilitate coping.}

§ 3.3.1 _{functional “performance dependence”. Therefore, role processes between organizations rely more on} Municipalities are expected to be able to influence boundary spanning administrators through interpersonal and inter-organizational relations as a substitute for a clear formal hierarchical structure. § 3.3.1 I expect that the model of Kahn et al. (1964) may need to be adapted when used in a complex

collaborative context to explicitly incorporate external role senders apart from internal role senders. § 3.3.2 Individual boundary spanning representatives rely more on personal sources of power and influence

rather than formal authority (e.g. trust).

§ 3.3.3 Communication of expectations, as well as feedback, of both internal and external role senders is vital for the effective performance of boundary spanning activities.

§3.4.4 Participation in decision making is expected to be an inadequate substitute for autonomy in boundary spanning activities.

4) RESEARCH METHODS

Nearly all prior research on role ambiguity and conflict has been quantitative and focused on correlation analysis. In order to get a better understanding of how the extended role episode model of Kahn et al. (1964) can be applied to illustrate the perception of boundary spanners on how antecedent factors of role conflict and ambiguity influence their performance in inter-organizational activities, I use a primarily qualitative approach with mere support from an in-depth questionnaire.

4.1 Data collection

interviewed for the WIDS project (Van Ees & Six, 2006), as well as this study, were also used, creating a set of 24 interviews in total.

The three water boards have experienced problems in collaborating with municipalities and wanted to learn from these events. A written research proposal was send to the three department heads, each in a different water boards, explaining the background and purpose of the study. Furthermore, interview topics and requests for secondary data were presented. Cooperation was confirmed over the phone. The interviewees were, based on requests, selected by the department heads. The common element of the interviewees I requested to interview was that they are people that have many contact moments with outsiders and that these contacts are important to performing their own jobs. These are similar to the two dimensions of boundary relevance Kahn et al. (1964) distinguished.

Due to differences in the organizational structures of water boards, interviewees were either from the strategy and planning department, external policy department or from one particular district. Before the interviews were held, each participant received a cover letter explaining the background and purpose of the study, excluding expectations. Individual confidentiality was guaranteed. The semi-structured interviews lasted approximately 90 minutes. The topics were presented in common order of discussing in Table 2.

TABLE 2

Topics discussed in the interviews

Topics discussed

Introduction & background of research Background of interviewee and activities

Role in boundary spanning activities

Identifying role senders (including guidelines, laws etc.) Similarity of role expectations and compatibility of role pressures

Role conflict and ambiguity experienced Coping patterns & Feedback

Power, influence & dependence between interviewees and role senders Changing role and complexity in inter-organizational activities

Organizational structure & training/development Organizational culture

After the interviews, a questionnaire was distributed to get more information on interviewees’ perceptions of 1) the organizational factors formalization, autonomy and hierarchy, 2) the organizational culture and leadership, 3) the role conflict and ambiguity experienced.

Between one and five weeks after the interviews were conducted, seven phone calls of approximately 15 to 20 minutes were made. The objective of the calls was primarily to clarify answers that were not clarified before or raised due to the questionnaire. Inconsistencies between answers of 1 interviewee led on one occasion to adapting questionnaire results.

In the water management domain, municipalities and water boards frequently cooperate. For the 21st century, the importance of these contacts for the effective performance of water boards is expected to grow, because water boards more often require municipalities to share goals in order to meet their own (Tielrooy et al, 2000). Interviewees were asked to focus on their cooperation with municipalities, because municipalities are thought to be the most important partner of water boards.

4.2 Measures

All 13 water board interviewees returned the questionnaires, of which ten were complete. In four sections of the questionnaire, administrators were asked to rate a number of questions on a six-point ordinal scale ranging from “strongly disagree” to “strongly agree”. In section on leadership style and organizational culture, the respondents were asked to allocate 100% over four categories.

Role conflict and ambiguity measures were developed by Rizzo et al. (1970). Their questionnaire is applicable to all positions in which incumbents enjoy some degree of autonomy. According to Jackson & Schuler (1985), 85% of all role research studies use their scale. Most items for role conflict and ambiguity were derived from this questionnaire and supplemented with water board specific questions about the transparency of the law and their position in inter-organizational projects.

The questions on the organizational structure consist of two components. All items from the modern and post-modern forms of organization (Williams, 2002) were included, as well as items often used to organic-mechanistic dimension (Burns and Stalker, 1961; Lawrence and Lorsch, 1967), with emphasis on formalization (House & Rizzo, 1972).

Secondary sources used in this research consist of, next to ten interviews of WIDS project (Van Ees & Six, 2006), evaluations of departments and yearly departmental plans, including development programs. The interviews were used to corroborate findings on inhibiting factors boundary spanners in water boards mentioned.

The yearly departmental plans were requested, because they present written guidelines on the goals and role of the department. I used this information to corroborate statements and research findings across multiple data sources over the development focus and programs for competencies development. This information was also used as input for interviews to ask questions related to the transparency of (and conflicts in) organizational goals. One department-wide evaluation report included 24 interviews and questionnaires that provided a very useful insight from a change management perspective on interviewees’ perception of the strategy department. Specifically, six municipality representatives showed their concerns. Organization-wide water governing plans include a vision and mission section, which was expected to provide more insight into the position of the boards towards cooperation and the importance boundary spanning activities. Many other water management specific documents, such as task descriptions, policy documents, departmental guidelines etc. were used as background information.

5) RESULTS

In this section, we examine the relationship between organizational, personal, and interpersonal factors and role performance to gain more insight into the process of how these factors in role theory affect role performance of boundary spanners in water boards.

who says “I would like to present myself as an advisor”, also says that “Our primary role is still to grant permits”.

Two other interviewees argue that, “Because we are both controllers as advisors we are perceived as a controller rather than a partner”. Three interviewees argue that although they have an advisory role, partners say “we do it, because the water board says so”. This finding is corroborated by the interviews with municipality representatives. One says “Advisory? Yes, but in very demanding way”. His superior stated, “Three of my people interpreted it as a demand” (…). Responsibilities were unclear (…). At one point I explicitly asked, do you have an advisory or a decisive role.”

In two other projects involving two other water boards, responsibilities about who was financially responsible was unclear. In the first example, as one interviewee argues, “We put too much effort in determining who was responsible” (…) Everyone engaged in a political process underneath collaborative efforts”. In the second project, as one interviewee argues “the water board didn’t take financial responsibility. (…) If we had known that beforehand, we would have operated very differently.”

Note though that these interviews with municipality representatives were all held in a context of projects gone badly. Furthermore, three of the seven municipality representatives had positive remarks as well, such as 1) “I also worked together with the water board on a water plan, in which I met different representatives and the cooperation was very different, 2) “Water board now have young people who put more emphasis on their advisory role. (…) Maybe that’s evolution, that nowadays you have to work together”, and 3) “My view of the water boards is changing. (…) I see a lot of new people in the water boards that have more feeling for water in the city”.

TABLE 3

Answer to questionnaire questions used in the results section

Question as posed in questionnaire Average Responses (N)

I have sufficient financial autonomy in inter-organizational

projects 2.6 12

I have sufficient process autonomy in inter-organizational

projects 5.4 9

In my line of work I sometimes comes across ambiguous laws 5 10

Guidelines from the government are clear 2.5 10

Internal decision making is based on negotiation and consensus

rather than hierarchy and power 2.9 12

Our policies and work methods are explicitly formalized 2.9 12 My responsibilities in inter-organizational activities are

explicitly formalized 2.50 12

We try to standardize as much services/processes as possible 4.5/ 2.0* 8 / 3* My work requires innovative and creative solutions to solve

non-routine problems 5.3 12

The influence I have in inter-organizational projects varies 5.2 11 I adapt my attitude and behaviour to the role I have in projects 5.0 11

In general, there is a lot of trust inter-organizational projects

between partners 4.0 12

Legend: * = 3 interviewees from one water board said 2, while 8 interviewees from two other water boards averaged a 4.5.

5.1 Organizational factors

5.1.1 Formalization

their abilities, even though they know that in theory it should not matter whether a colleague or he/she performed the activities.

The tendency towards formalization is said to be high in two water boards (4,5 on average), and low in one other (2 on average). The difference is consistent with the fact that the latter is currently reorganizing. Reorganizing, rather than formalizing, is a priority at the moment. All interviewees agreed that their job require innovative solutions to non-routine problems (5.3 on average). One interviewee reported that formalization conflicted with creativity when too much was formalized. Three interviewees in water boards A and C argued that there should be general guidelines for consensus on how to do things, but leave some room for incorporating specifics of a project.

5.1.2 Autonomy

I differentiated between the autonomy on the process of performing activities and financial autonomy. Financial autonomy refers to an administrator’s authority to make decisions concerning the means to achieve organizational goals. Note here that autonomy in the process of cooperation does not require authority. E.g. “I can cooperate the way I see best fit” refers to formalization of processes and behaviour rather than authority.

The average score for “sufficient financial autonomy” was 2.6, while the average score for “sufficient process autonomy” was 5.4. Interviewees responded on the question whether negotiation and consensus plays a more important role in internal decision making than power and hierarchy the average score of 2,92. A common finding was that while responsibilities are often delegated, the authority to make decisions with financial or strategic implications is.

Two interviewees argued that they had little but also did not require much financial autonomy in their activities. One interviewee argued it wasn’t necessary to successfully perform her liaison role between an internal department and outsiders, while the other required clarification of the organizational strategy rather than authority. A notable point is here that three interviewees (two in water board A, one in C) argue that authority is only required for small projects, because “ As long as you have enough time to report back, it isn’t such a problem”. Furthermore, it seems reasonable to assume that partners do not expect non-managerial boundary spanners in water boards to make decisions with large strategic or financial implications in one meeting anyway. Conflicts are therefore created by a combination of externals’ ambiguity on the limited financial authority of boundary spanning administrators in water boards and internal lack of authority. Furthermore, since these two interviewees rated a 4 on this particular item in the questionnaire, the “insufficient financial autonomy” result is more likely to be underestimated than overestimated for interviewees who do value and require this authority.

As a coping technique, three interviewees use a board member as a sounding-board on whether certain recommendations are expected to be acceptable. Two interviewees explicitly mentioned a lack of authority to frustrate the process of cooperation, because they have little bargaining power to reach agreements. Two other interviewees stated that it took a lot of time to write proposals of which they knew were almost certain to be accepted.

5.1.3 Organizational fit

The organizational structure of organizations defines the position of the role senders and focal persons. A boundary position was expected to and is found to be different from a non-boundary spanning position in terms of their role senders and pressures placed upon the positions. First, administrators in boundary positions argue that they require cooperation with outsiders to be able to perform their activities successfully. A representative quote is “If I lose contact, the organization loses contact” and “If the municipality is not interested, cooperation becomes futile quickly”. 11 interviewees mentioned that a boundary spanner relies heavily on trust and the interest of others, because they often have no formal position to achieve optimal results; goals can only be reached independently at higher costs. Therefore, boundary spanners are “not able to dictate to the outside world”

board boundary spanners is to explain the situation to municipality representatives through communication and find solutions that are mutually satisfactory. 10 of the 14 interviewees stated that the most important competency for them was the ability to understand the position of partners to be able to integrate objectives.

Organizational fit was defined as the coherence between organizational goals, leadership and culture of water boards. The culture of the boards, particularly the general board, is by all interviewees said to be “conservative”. A majority found their focus on cost reduction too conservative and conflicting with organizational goals in yearly and long-term plans. Results of part of the questionnaire concerning culture (Cameron and Quinn, 1999) are presented in Table 4. Note though that in these two organizations, there is an external/strategy department, whereas in the other water board, a divisional structure is used.

All 13 interviewees agree that they themselves have become more pro-active in their boundary spanning activities. This entails that they frequently and actively seek communication and cooperation with outsiders in order to achieve better results. In water board B, secondary sources support the frequently mentioned departmental focus on communication and cooperation. Two examples are training on “influencing without formal authority”, and a culture meeting about “the change from an inward focused organization towards a more external oriented organization.”. Interviewees in water board B also argued that their department is less pragmatic, more academic, and focused on cooperation.

TABLE 4

Water board A Water board B Water board C

Market culture _{21,3 % 23,8 % 36,7 %}

Clan culture _{8,8 % 17,5 % 20 %}

Hierarchy culture _{47,5 % 42,5 % 20 %}

Ad Hoc culture _{27,5 % 16,3 % 23,3 %}

TABLE 5:

Water board A Water board B Water board C

Facilitator/Mentor _{50.0 % 36,7 % 16,7 %}

Monitor/Coordinator _{19,4 % 41,7 % 23,3 %}

Competitor/Producer _{11,9 % 13,3 % 36,7 %}

In water boards A and B, with a external/strategy department, the interviewed administrators indicate that there is a primarily hierarchical culture (47,5; 42,5) in the organization, while in the more “conservative” water board C, there is a focus on “market culture” (36,7). A likely explanation for these differences is found in the annual departmental development plans, in which the focus on water board C, in contrast to water board A, and especially B, is on operating “market conform”. Furthermore, interviewees also argue that without doubt, the primary focus of their boards of control and management is cost reduction. This finding is supported by Table 5, which clearly shows that their manager is perceived to be a competitor/producer. Other signals of a cost focus that were provided by interviewees, especially in water board C, but also in A, is a vacancy stop, use of temporary labour, and the amount of time available for activities that did not directly present results. These direct effects of cost focused general boards were perceived as not supporting the change towards a more pro-active organization in inter-organizational cooperation.

All 13 interviewees mentioned a cultural gap between them and their boards of control in the sense that they found their boards of control, especially the daily board, to be more conservative. In water boards A and B, some interviewees believe that the organizational direction is not clear enough, because the emphasis of daily management is on development, while the message from the general board is cost-focused. I also found a development focus written in the yearly departmental plans, which is clearly more enthusiastic than primarily the general board is. An interesting finding is that in water C the yearly departmental development plan has ambitious development plans as well, but are according to the interviewees, not (properly) executed. Finally, in an evaluation report of the strategy department of water board B, it is argued by general board representatives themselves that although they want to be more proactive in the organization, “Some general board members are not present at meetings through which they, by asking questions and objections, cause delays in a later stage of the process.”

5.2 Personal factors

5.2.1 Competencies

factors for administrators in boundary spanning positions. The one interviewee that indicated that expert-knowledge in his particular speciality was most important, spent the least time with externals (10%). The most important competency is, according to interviewees, the ability to integrate your objectives and that of partners into one plan by understanding their pattern of thinking. This is similar to: “being sensitive to and skilled in bridging interests” (Webb, 1991), “understanding another’s organization and to make a real effect to empathize with” (Trevillion, 1991), and “appreciating the interdependencies in ones job” (Williams, 2002). Also, the ability to communicate clearly and effectively (Williams, 2002), is mentioned 8 times by interviewees.

TABLE 6

Competencies found most important by interviewees (n = 14)

Competency Times

mentioned

• The ability to integrate your objectives and that of partners into one plan by being able to think the way partners do through understanding their positions in inter-organizational activities

10

• Communication skills 8

• Strategic insight into opportunities at hand and the chances of success for realization when future steps are made, including the possible that there will be support from the board(s) of control.

3

• Innovativeness and the ability to translate strategic goals into practical operational goals as important competencies

3

• Specific knowledge on your work area in water management 3

• The ability to refrain from pseudo-activities and focus on the core

business and responsibilities of water boards 2

• Ability to generate trust 2

5.2.2 Personality & Attitude

Three interviewees explained that keeping an open mind about alternative solutions/opportunities and being friendly but assertive was the most successful in organizational activities. Furthermore, interviewees stated that the role of the water board determines the role of the boundary spanner and the influence he/she has (average of 5,2) and how he/she behaves (average of 5). When the water board is more interested in terms of opportunities, a more pro-active role is said to be required.

An internal locus of control does seem to be beneficial for boundary spanning activities. Three interviewees mentioned that in their position you have to be tolerant for ambiguity and seek actively for information. It is not surprising that interviewees in water boards A and B argue that their supervisor has a facilitator/mentor role (see Table 5), because they work rather independently, seeking information and opportunities themselves due to the complexity of projects. When facing ambiguity, interviewees state they directly turn to person creating the ambiguity, internal or external. All interviewees say they primarily turn to peers to provide specific information and to their managers for information concerning the process of cooperation. Interviewees also argue they tend to be better informed than their superiors about specific projects. An important goal for the majority of interviewees is to “exert influence”. The best way to do so is by providing input on water preferences in an early stage of a planning process, where changes can be made relatively easily.

5.3 Interpersonal factors

5.3.1 Interpersonal relations in an internal and external context

on boundary spanning administrators that are different from internal role senders. Therefore, I argue, they should be separated in the role theory model of Kahn et al. (1964).

While investigating whether internal role episodes are different from inter-organizational role episodes, I found that high distance role senders, both internal and external, account for most of the role ambiguity and conflict in all three water boards.2 General board members and water board representatives do not always have adequate conceptions of the boundary spanning administrators’ role. In all three water boards, the majority of interviewees argue that stating goals is one thing, but the real challenge lies in making the resources (time and money) available to achieve those goals.

5.3.2 Power, dependence & trust

Power was defined as the ability to influence others to believe, behave, or act. Municipalities and water boards frequently cooperate and compromise. As a result, they have a long history together and often rely on cooperation in projects to achieve their own organizational goals. All but one interviewee explained that informal power played a significant role in their inter-organizational activities. Basically, municipalities rely on water boards for permits, and water boards also rely on municipalities for reaching their own goals, which municipalities also value to various lengths. Therefore, these organizations are mutually dependent and cannot present their demands unilaterally. Hence inter-organizational dependence between water boards and municipalities undermines formal power and stimulates reliance on trust for cooperation. Interviewees state that they do want to use their formal power, because it harms their relationship with partners. In other words, administrators choose to rely more on cooperation, because cooperation often allows both organizations to achieve organizational goals. Especially expertise, legitimacy and trust are often mentioned as important substitutes for the coercive power they can derive from their formal legal position.

Inter-organizational relations were found to influence interpersonal relations. Five interviewees mentioned that 1) the role of the water board, 2) the role of the municipality, and 3) the overlap in interests as the most important determinants of influence. Three interviewees argued that in order to reach the targets for water quality, the water board may need to make more resources that can be used as bargaining power.

In a meeting with several water board administrators, they indicated that trust was one of the most important success factors in inter-organizational activities. Interpersonal trust has

2

grown in importance as a substitute for formal power in the (inter-)organizational hierarchical structure. When I asked the interviewees to give their opinion as to what were important factors that determined the influence they had in inter-organizational activities, five interviewees indicated that trust was most important. Currently, interviewees are moderately positive about the amount of trust between partners in inter-organizational projects (average of 4, cf. Table 3). Trust allows parties to look beyond their formal positions and allows cooperation. Furthermore, trust increases the receptiveness of municipalities. When the water board is not financially contributing and has no coercive power or decides not to use it, administrators in boundary spanning roles rely on expert and legitimate power. Trust is found to be a mediator between these two sources of power and the influence on a partner. Surprisingly, although the existence of trust is considered extremely important, the ability to generate trust was only mentioned twice as a competency (see Table 6). The former is in sharp contrast with prior boundary spanning literature, which emphasis managing trust and building sustainable relations as a crucial part of a boundary spanners job.

There appears to be a complex relationship between autonomy and trust. Problems occur when administrators in water boards find themselves in a position where they have to make compromises and lack the autonomy to engage in reciprocal risk taking. As expected, boundary spanners are then confronted with the accountability interface between their role as organizational representative and that of a partner in a collaborative context. This lack of authority is found to inhibit trust building when conflicts occur as a result of ambiguity on expectations. Two municipality representatives, as well as two water board administrators (in water boards A and C) explicitly mentioned this causality.