Module 4 through 7 has an associated group discussion that should focus on discussing the course content for that Module. Each discussion will span the two-weeks of the Module. Each s

©The Author 2017. Published by Oxford University Press on behalf of the Public Management Research Association. All rights reserved. For permissions, please e-mail: [email protected]

1

Journal of Public Administration Research And Theory, 2017, 1–18 doi:10.1093/jopart/mux026

Article

Article

Picking the Team: A Preliminary Experimental Study of the Activation of Collaborative Network Members Chris Silvia

Brigham Young University

Address correspondence to the author at [email protected]

Abstract

Among a collaborative leader’s most important duties is selecting a collaborative partner. Numerous perspectives, including resource dependency theory, institutional theory, transaction cost theory, and personality typologies, have been used to help explain this decision. Clearly, a collaborative leader would desire to work with an individual who has access to needed resources, has a personality that fits the network, and is familiar. However, such a perfect partner does not often, if ever, exist. Therefore, a collaborative leader must make trade-offs between the issues of resource access, personality, and familiarity. Using an experimental design, this study explores how collaborative leaders make these trade-offs when considering potential collaborative part- ners. The findings suggest that while prospective partner personality may be the most significant driver of the partnership decision, it is actually the combination of factors, especially personality and resource access that interact to determine partner desirability.

Introduction

Over the last few decades, government leaders have increasingly utilized collaborative approaches to prob- lem solving and service delivery (O’Toole 1997). As McGuire and Silvia (2010) commented, “the public manager of the current era must regularly and skillfully navigate a multitude of actors and programs in the intergovernmental system,” and “there is little doubt that interdependence and interconnectedness charac- terize the intergovernmental environment of today’s public organizations” (279). However, despite the ubiquity of collaborative arrangements, the practice of collaboration is not easy. Collaborative leaders and the networks they lead face many challenges, such as knowledge gaps, legal barriers, resource scarcity, exer- cise of agency negative power, and political opposition (Agranoff 2012). Central to a collaborative leader’s role is the selection of the people and resources that the network needs to achieve its objectives (O’Leary

and Vij 2012). A collaborative leader must determine the discrepancy between what the network needs in terms of people, perspectives, and resources and what it currently has. Given that the rationale for collabo- rating is often the realization that a problem “cannot be solved—or solved easily—by a single organization” (McGuire 2006, 34), the decision regarding with whom to collaborate is not trivial. Graddy and Chen (2009) suggest “three factors that [they] believe explain the choice of partners in collaborative arrangements: pro- grammatic needs that promote resource exchange; organization legitimacy goals; and efforts to reduce the transaction costs associated with partnership for- mation and management” (55). By identifying and recruiting partners, a collaborative leader can amass the resources, garner the information, and assem- ble the people that individual organizations lack, yet require, to address the issue that brings them together. This accumulation and retention of “resources like

Journal of Public Administration Research and Theory, 2016, Vol. xx, No. xx2

money, information, and expertise can be the integrat- ing mechanism of networks” (Agranoff and McGuire 2001, 298) that maintains a sense of interdependence among the collaborative parties and engenders stabil- ity within the network membership by incentivizing participation and fostering the network’s ability to achieve its mission and goals (Ansell and Gash 2008).

Many scholars have also argued that the incorpora- tion of the right partners is a key to success (Graddy and Chen 2009). Central to this idea is that who is asked to participate in the collaborative is as impor- tant as what the potential partners can contribute in terms of resources. This is because, while collabora- tive partners each represent their own home agencies or interest groups, it is the people, not the organiza- tions that actually collaborate. The convener of the collaborative must carefully consider the personalities of each partner candidate. Hackman (2011, 155) pos- ited that “60 percent of the difference in how well a team eventually performs is determined by the quality of prework,” including determining the group com- position. Identifying the “right” people is important because of the shared leadership arrangement inherent in many collaboratives. Perhaps more so than in a hier- archical structure, “every man in a group is to some extent a leader in so far as every man has some effect upon the syntality of a group” (Cattell 1951, 25). In other words, the collaborative’s personality is a func- tion of the personalities of the individual members of the collaborative and the results of the collaborative effort “depend on the team members’ characteristics” (Vincentini and Boccardelli 2014, 40). Each partner’s contribution to the overall success of the collaborative is not only a function of the resources that they con- tribute, such as money, information, reputation, etc., but also each individual’s attributes.

Collaborative arrangements are not without their skeptics. A  common caution from practitioners and scholars is to collaborate only when necessary because collaboration is a “seriously resource-consuming activ- ity” (Huxham and Vangen 2005, 13)  that requires a great deal of time and energy to make it successful. In their examination of the impact of collaborative lead- ership behaviors on network effectiveness, McGuire and Silvia (2009) found that engaging in framing behaviors, including those leadership behaviors that focus on establishing agreement regarding individu- als’ roles and the operating rules of the network, were negatively related to network effectiveness. They argue that if the leader must continually spend time getting everyone on the same page, then the members lack the unity necessary for success. The lack of agreement means that the network must keep its focus on deter- mining how to work together instead of actually work- ing together toward accomplishing their mission and

goals. “Organizational performance is an outcome of how well . . . interdependent elements are aligned . . . [and] are working in concert to attain specific goals that ultimately help the organization fulfill its mission” (Daft 2008, 216). One way that managers attempt to decrease these transaction costs is to work with those with whom they are familiar. “Members who are familiar with one another and with their work con- text are able to settle in and focus on the work rather than waste time and energy getting oriented to new coworkers or circumstances” (Hackman 2011, 62). Thus, working with familiar individuals is not only easier, since they are a known quantity, but also leads to greater efficiency and effectiveness.

Recently, there has been an increase in scholarly interest regarding collaborative partner selection (e.g., Berardo and Scholz 2010; Calanni et al. 2015; Feiock, Lee, and Park 2012; Graddy and Chen 2009; Ryu 2014). Each of these studies has added much to the conversation regarding this important management topic. This article seeks to build on these earlier studies by examining the trade-offs collaborative leaders make when selecting pro- spective partners. As discussed above, partner selection is influenced by the potential partner’s access to resources, potential to increase the collaborative’s legitimacy, famil- iarity among the current members of the collaborative, and his or her personality. Certainly, a collaborative leader would like to include a person who represents an organization that has needed resources, has a personality that fits the network, and is familiar. However, since the perfect collaborative partner likely does not exist, trade- offs must be made. This article strives to discover how public managers make these trade-offs when considering a potential collaborative partner. Thus, the research ques- tion explored in this article seeks to determine which mix of benefits (i.e., resource exchange, associational legiti- mization, transaction cost reduction, and personality type) is seen as providing the best value when confront- ing the inevitable trade-offs that must be addressed when selecting partners.

Literature Review

This study brings together a number of different lit- eratures as a foundation for the examination of how collaborative partners are selected. In so doing, there are a number of terms that will be used in this article that either may not be consistently defined within or across the fields from which they are taken. For exam- ple, a term such as “collaboration” likely means differ- ent things to different people based upon the field of scholarship. Therefore, it is essential that these terms be defined for the purpose of this study.

Consistent with definitions used by Agranoff and McGuire (2003) and Huxham and Vangen (2005), the

Journal of Public Administration Research and Theory, 2016, Vol. xx, No. xx 3

term “collaboration” will be used in this study to refer to the process by which two or more entities work across organizational boundaries to achieve something that they could not accomplish alone. Networks, on the other hand, are structures. O’Toole defined networks as “structures of interdependence involving multiple organizations or parts thereof, where one unit is not merely the formal subordinate of the others in some larger hierarchical arrangement” (1997, 45). Thus, the term “network,” which will be used synonymously in this study with terms like “a collaborative” or “a col- laborative group,” refers to the organizational struc- ture within which the process of collaboration takes place.

The literature differentiates between types of net- works based upon their function. For example, knowl- edge networks (e.g., Eglene, Dawes, and Schneider 2007) form as a forum within which to share infor- mation, governance networks (e.g., Klijn and Skelcher 2007) form to make and implement policy, and service delivery networks (e.g., Romzek et al. 2014) form to facilitate the rendering of public services. However, regardless of the function of the network, they must all find and incorporate new members to ensure that the network has the right people and resources to execute its mission.

The literature also differentiates between types of networks based upon their structure. Perhaps the most notable example is Provan and Kenis’ (2008) typol- ogy consisting of three forms: network administrative organization, lead agent-governed, and participant- governed. In these structures, the network adminis- trator, lead agent, and individual network members, respectively, are all tasked with identifying and recruiting new members as the circumstances require. Therefore, in the context of this current study, the dif- ferences between network functions and structures are not critical.

Others differentiate networks based upon their institutional constraints. “Collaboration can be either formal (mandated by the state) or informal, involve many organizations or few, it can be vertical and/or horizontal, and it can be intra- and interorganiza- tional” (Smith 2009, 1). Constraints such as these play a significant role in the partnership decision. For exam- ple, there are many situations where collaboration is not only mandated, but the collaborative participants are also mandated. Further, all collaborative arrange- ments are different and thus will operate by different rules. While these are important factors in collabora- tion, institutional factors that constrain partner choice are not explicitly included in this study. Instead, a basic framework for partnership selection based upon col- laborative style, resources, and familiarity is proposed and tested. By removing these contextual variables,

the underlying partner selection logic of the decision maker can be identified more clearly. This study is a first step and can serve as a scaffold upon which future studies can build to understand how the findings pre- sented here can be applied to specific institutional arrangements.

For purposes of this article, an individual within the network that is considering incorporating a new mem- ber will be referred to using the term “collaborative leader” and the new member will be referred to using the term “potential partner.” The term “collaborative leader” does not indicate that this individual exerts complete control over who participates in the network. As Feyerherm (1994) concludes in her study, in a col- laborative setting, leadership is often exhibited by both the acknowledged leader as well as by the other mem- bers of the collaborative. Instead of a single collabora- tive leader, the leadership role is shared by multiple, and often all, members of the network (e.g., Ansell and Gash 2008; Crosby and Bryson 2005). This diffusion of leadership functions means that all members can, and likely do, play the role of a leader and that those who are recruited as partners are both future members and future leaders of the group. The potential partner is an individual that represents his or her home organi- zation in the network.

There are numerous perspectives that can speak to collaborative partner selection. Those considered in this study are resource dependency theory, institutional theory, transaction cost theory, and syntality/person- ality. Each perspective offers a different rationale for the inclusion of a potential partner. Importantly, these perspectives are not mutually exclusive, since each pro- spective partner will have a unique mix of access to resources, legitimating power, familiarity among the current collaborative partners, and personality type. Thus, the aim of this study is not to test these four per- spectives against each other, but rather to identify the trade-offs among them that dictate partner selection.

Resource Dependency Theory Arguably the most common perspective (Hu, Khosa, and Kapucu 2015) used in the context of collabora- tive partner selection is resource dependency (Pfeffer and Salancik 2003). A  single organization’s inability to address an issue on its own often results from a lack of necessary resources. Thus, when a network lacks the resources it needs to accomplish its goals and mis- sion, it is often incumbent upon a collaborative leader to identify and recruit partners that have access to those resources. These partners work for organiza- tions that can provide the network with the desired resources. A collaborative leader, therefore, may pur- sue a collaborative partnership as a way to manage uncertainties in their network’s resource environment

Journal of Public Administration Research and Theory, 2016, Vol. xx, No. xx4

(Gray and Wood 1991). Based upon her case study work, Cigler (2001) identified a set of nine precon- ditions, including capacity building, that help explain the decision to pursue a collaborative venture. Cigler’s work “could not uncover even one organization that had emerged without some type of capacity-building assistance external to the organization” (78–9). As some Pentagon planners have said, “a vision without resources is a hallucination” (Friedman 2009, 205). Thus, resource dependency theory would suggest that the desirability of a collaborative partner is based on their access to the resources needed to successfully meet the network’s objectives that are not currently possessed by either the collaborative leader’s home organization or by the other members of the network (Bardach 1998; Gray and Wood 1991; Malatesta and Smith 2014).

Institutional Theory Another theory that is frequently used to help explain collaboration is institutional theory, which suggests that the selection of a collaborative partner is based on the network’s strategic decision to improve its reputa- tion, image, or prestige among its own membership and stakeholders (DiMaggio and Powell 1983;Graddy and Chen 2009). “Collaborative projects often need broad public support and/or public legitimacy because they normally represent the facilitated sharing of power and resources from many agencies and affected persons” (Agranoff 2012, 12). However, collaborations are not automatically granted legitimacy (Bryson, Crosby, and Stone 2006). Instead, the collaborative must build and maintain legitimacy, an objective that can potentially be achieved through the efforts and reputation of col- laborative partners.

The reputation of collaborative partners has a num- ber of distinct advantages for the network at various phases of the network’s life cycle. Human and Provan (2000) identified three distinct aspects of network legitimacy: as a form, as an entity, and as an interac- tion. Since the collaborative partners are the face of the collaborative—both within the network and to exter- nal stakeholders—prestigious partners can help the network achieve legitimacy in all three areas.

In terms of building legitimacy of the network as a form, having a highly visible individual involved can help establish the network as an acceptable form of organizing for the purposes of solving an intractable problem. This is particularly the case in the beginning stages of the collaborative’s existence because it helps with recruiting collaborative partners (Linden 2010). In addition to helping establish the credibility of the network as an organizational form in the eyes of poten- tial collaborative partners, networks must also estab- lish the credibility of the network as a form in the eyes

of external groups and stakeholders, including poten- tial funders (Human and Provan 2000). This point was illustrated in Crosby and Bryson’s (2005) discussion of the 2002 Vital Aging Summit in Minnesota, in which “members of the summit planning committee used their personal connections and credibility with foun- dation directors to win donations” (303).

Once the legitimacy of the network as a form of organ- izing is established, the network must establish its own legitimacy as an entity. Of his seven key collaborative factors, Linden (2010) identified the inclusion of “some- one with credibility and clout . . . [as potentially] the most important” (49). Linden’s argument for this assertion is essentially that a champion can help the network gain legitimacy as an entity both among the network mem- bers and their stakeholders, thereby gaining their sup- port. Demonstrating the legitimacy of the network and actively mobilizing the support of both members and stakeholders of the network have been found to be inte- gral to the success of the network (McGuire and Silvia 2009). This was illustrated during the establishment of the Hennepin County African American Men Project, where the collaborative leaders sought out “African Americans who had rich connections and credibility in their own ethnic community,” resulting in a collabora- tive “that had prestige in the eyes of many stakeholders” (Crosby and Bryson 2005, 66). As Crosby and Bryson remark, “solutions gain prestige through attachment to respected people, institutions, or processes” (248).

The final dimension of legitimacy that is important for collaborative leaders to consider is that of the network as an interaction. “The interaction process itself [has] to be legitimized so network members [will] be willing to work together to build and maintain the levels of involve- ment and norms of cooperation that would be critical for sustaining the network (Powell 1990)” (Human and Provan 2000, 340). This is akin to McGuire and Silvia’s (2009) findings regarding mobilizing and synthesizing, both of which were found to be positively associated with network effectiveness. Mobilization behaviors are those that “develop support for network processes from network participants” and involve “establishing and maintaining [the network’s] legitimacy” (39–40), and synthesizing behaviors are those that “create and main- tain trust among network participants as a means to build relationships and interactions that result in achiev- ing the network’s purpose” (40). Partner trustworthiness has been cited as being especially important in situations where there is increased commitment risk (Feiock, Lee, and Park 2012) or where there is low social trust and a lack of potential collaborative partners with whom one is familiar or with whom one has an existing personal relationship. In such contexts, collaborating with a repu- table party “could provide a ‘guarantee’ of reliability” (Koljatic, Silva, and Valenzuela 2006, 65).

Journal of Public Administration Research and Theory, 2016, Vol. xx, No. xx 5

Transaction Cost Theory As mentioned previously, the selection of a collabora- tive partner is frequently driven by the identification of a need that is not being met by the current compo- sition of the network. Often it is familiar individuals who come to mind when deciding whom to recruit to fill that deficiency. From a transaction cost perspective, this makes sense: recruiting a known entity decreases the search and information cost of identifying others with the personalities and capabilities of the familiar individuals. As also mentioned earlier, uncertainty is a concern for a collaborative leader, both in terms of whether to and with whom to collaborate. One way to reduce this uncertainty is to collaborate with some- one the leader knows and has worked with in the past (Hinds et al. 2000). Hinds et al. (2000) identified two key mechanisms by which this decision occurs. First, past experience is the best predictor of future perfor- mance. Second, less time is required to socialize and bring a familiar partner up to speed (Cohen, Ledford, and Spreitzer 1996). There is also evidence that work- ing with familiar individuals increases productivity and decision-making effectiveness (Sawyer 2007). Sawyer (2007) identified familiarity as being particu- larly helpful for tasks requiring “problem solving crea- tivity” (52) and as one of the ten conditions that foster group flow, or “a peak experience, a group performing at its top level of ability” (43). Familiarity often arises from a history of successful past collaboration, which has been found to engender trust because partners are better able to predict each other’s actions (Gulati and Sytch 2008). Familiarity decreases the uncertainty of a partnership and helps collaborators to establish a “common language and a common set of unspo- ken understandings” (Sawyer 2007, 51) and “mutual understanding, internal legitimacy, and shared com- mitment” (Emerson, Nabatchi, and Balogh 2012, 14).

Personality The final major factor that has been used to explain col- laborative partner selection is personality. Among the resources that a collaborative has at its disposal is the individual characteristics of its members. While little research has been done regarding the impact of person- ality traits in the area of collaboration, there is a rich literature on this subject relating to group dynamics and group effectiveness (e.g. Bell 2007; Kramer, Bhave, and Johnson 2014; LePine et al. 2011). The major premise behind this stream of research is that the “personality factors of team members are characteristic patterns of thinking, feeling, and acting that should affect team per- formance through a variety of processes ranging from how team members approach task completion to how team members interact with one another” (Bell 2007, 597). Personality and individual characteristics are

thought to be particularly important in group dynam- ics because they “are not only directly relevant to the task-focused contributions individual members make to team outcomes, but they also influence the manner in which members react and relate to each other in the course of performing work together as ongoing units of work” (LePine et al. 2011, 312). Thus, there is likely an interaction between the partners’ personalities that not only impacts the effectiveness and efficiency of the col- laborative’s work but also the group dynamic in terms of trust and mutual understanding.

Among the myriad personality typologies are Parker’s (2008) team player styles. Parker’s typology was developed using data from approximately 3,400 individuals working in team, group, and collaborative arrangements. The styles include contributor (task- oriented), collaborator (goal-directed), communica- tor (process-oriented), and challenger (questioning). Parker (205–6) provides the following descriptions:

Contributor The contributor is a task-oriented team mem- ber who enjoys providing the team with good technical information and data, does his or her homework, and pushes the team to set high per- formance standards and to use their resources wisely. Most people see a contributor as depend- able, although they believe that at times a con- tributor may become too bogged down in the details and data or fail to see the big picture or the need for a positive team climate. Collaborator The collaborator is a goal-directed member who sees the vision, mission, or goal of the team as paramount but is flexible and open to new ideas, willing to pitch in and work outside his or her defined role, and able to share the limelight with other team members. Most people see a collab- orator as a big-picture person, but they believe that at times a collaborator may fail to periodi- cally revisit the mission, give enough attention to the basic team tasks, or consider the individual needs of other team members. Communicator The communicator is a process-oriented mem- ber who is an effective listener and facilitator of involvement, conflict resolution, consensus build- ing, feedback, and the building of an informal, relaxed climate. Most people see a communica- tor as a positive “people person,” but they find that at times a communicator may see the process as an end in itself and that a communicator may not confront other team members or give enough emphasis to completing task assignments and making progress toward team goals.

Journal of Public Administration Research and Theory, 2016, Vol. xx, No. xx6

Challenger The challenger is a member who questions the goals, methods, and even the ethics of the team; is willing to disagree with the leader or higher authority; and encourages the team to take well- conceived risks. Most people appreciate the value of a challenger’s candor and openness, but they think that at times a challenger may not know when to back off an issue or may become self- righteous and try to push the team too far.

There are a number of ways that personality could be theorized as affecting the partnership decision. If Wernerfelt’s (1984) broad definition of a resource as being anything that could be considered as a strength for an organization is applied to collaborative groups, it is easy to see how these collaborative role person- alities described above could themselves be seen as resources. Each role type has unique strengths that balance out the weaknesses of the others. Achieving a balance between the different roles could be important for the proper functioning of a group. Therefore, see- ing personality as a resource that can be incorporated into the network, a leader may purposefully collabo- rate with those unlike themselves or the other members of the collaborative in order to harness the comple- mentarity of the roles. Conversely, a leader may see a potential partner’s personality as impacting the trans- action costs of working together. Certain personalities are just more difficult to work with than others. Hence, leaders may purposefully avoid potential partners who are challengers, a personality type that has a reputa- tion as being difficult to work with (Clay-Williams and Braithwaite 2015). Finally, homophily would predict that collaborative leaders will choose partners that are like themselves because similarity not only increases the likelihood of collaboration (Feiock, Lee, and Park 2012) but also may spawn stronger relationships (McPherson, Smith-Lovin, and Cook 2001).

Data and Methods

Although experimental research designs are becom- ing increasingly more common across the social sci- ences, such designs remain relatively uncommon in public management research (Christensen et al. 2013; Margetts 2011). While novel research methods are interesting, the decision regarding which method to use must be based on the research question being asked. Since the objective of this study was to inves- tigate the trade-offs that a collaborative leader makes when selecting a collaborative partner, this article relies on a mixed-experimental design to uncover the causal relationships underlying this decision. The fol- lowing discussion highlights how this design meets the three criteria necessary to make a claim of a causal

relationship between the intervention and group differ- ences: an intervention, the measurement of outcomes,