Leiden University
The Role of
Interorganizational
Trust in Mandated
Policy Networks:
How does it affect network
effectiveness?
Erin Ellis, S1925237
MSc Public Administration
Public Management & Leadership
Supervisors: Dr. Petra van den Bekerom, Dr. Jelmer Schalk
March 13, 2019
Page | 1
Table of Contents
Introduction & Research Question ... 3
Theory ... 7
Networks ... 7
Networks Defined ... 7
Types of Networks ... 7
Network Effectiveness ... 8
Network Effectiveness Defined ... 8
Levels of Network Effectiveness ... 9
Benefits of Network Effectiveness ... 10
Trust ... 11 Trust Defined ... 11 Types of Trust ... 11 Trust in Networks ... 12 Benefits of Trust ... 13 Research Design ... 16 Case Selection ... 16 Operationalisation ... 17
Method of Data Collection ... 20
Method of Analysis ... 21
Empirical Findings & Analysis ... 22
Network Effectiveness ... 22
Objective Network Effectiveness ... 22
Subjective Network Effectiveness ... 26
Interorganizational Trust ... 28
Presence of Interorganizational Trust ... 29
Types of Interorganizational Trust ... 30
The Impact of Interorganizational Trust on Network Effectiveness ... 30
Conclusion ... 32
Bibliography ... 34
Appendices ... 37
Appendix A - Network Coordinator Interview Report ... 37
Page | 2 Appendix C – Codebooks ... 65
Page | 3
Introduction & Research Question
In 2014, a policy change was put into effect that made all primary and secondary schools in the Netherlands responsible for providing suitable education to every child, including those with special needs such as dyslexia arrangements, homework classes, and remedial teaching. Before this policy was put into effect, the so-called 'backpack system' was used. Under this system, funding was tied to each student, allowing them to fund and receive the education necessary for them outside of the regular education system (VNG, 2017; Schuman, 2007). This was deemed ineffective by policy makers and government officials alike because according to them, it caused many children to be labelled with special needs, even if this was not really the case. Additionally, the 'backpack system' was highly bureaucratic, the special education provided within the system was not always effective, and the costs of the system were continuously rising (VNG, 2017; Schuman, 2007). The Inspectorate of Education found that too much special education was being provided outside of the classroom and outside of schools with little to no way of measuring the results (VNG, 2017). The parents of children with special needs also admitted that they wanted special education to be provided in a less restrictive environment, and that not all students with special needs needed to be separated from the regular education system (Schuman, 2007). For these reasons, policy makers adapted a new approach to special education provision within the Netherlands.
In order to provide each child with the education that he/she needs, the new approach was a policy that grouped primary school boards into approximately 75 regional networks and grouped secondary school boards into approximately 76 regional networks. Each of the school boards within these networks agrees on an educational support plan based on what its schools can offer. The entire network receives its own funding and is responsible for the allocation of resources to provide quality special education. This way, students with special needs were able to return to the regular education system, and if one school board within a network is unable to provide a student with the education that he/she needs, the student is referred to another school board within the network that is able to do so.
Government has played a major role in initiating the formation of these regional networks. The school boards are incentivised to work together through government funding and administrative pressures, making them members of a mandated network. At the same time, it is a commonly accepted notion that network members cooperate and continue to cooperate based on some element of trust, especially in non-mandated networks where there are little to no incentives for cooperation provided by
government. Trust is important for the effectiveness of these types of networks since they lack a hierarchical structure and certain mechanisms, which leads to uncertainties, risks, and additional costs (Lambright, Mischen, & Laramee, 2010). Trust in some ways replaces these mechanisms and allows for the existence of reciprocal relationships, which are a necessary condition for the existence of networks and help to reduce the aforementioned uncertainties, risks, and costs.
With these regional networks in place for a few years already, it is interesting to discover what the perception of trust is within the networks, and how this affects the effectiveness of a network as a whole. We will do so by looking at one regional network, namely Samenwerkingsverband Passend Onderwijs VO 2801, or SWV VO 2801, as an in-depth case study. The network consists of 12 secondary school boards with 26 school locations. They are located in the regions of Voorschoten, Wassenaar, Leiderdorp, Zoeterwoude, Oegstgeest, Leiden, and Kaag & Braassem. The case will be described in further detail later in the Research Design.
Page | 4 We will be measuring the effectiveness of this network both objectively and subjectively, firstly, because this is an in-depth case study, and secondly, because measuring network effectiveness objectively is a large task, with challenges of selecting the appropriate performance measurement indicators in order to measure network effectiveness from different levels (Provan & Kenis, 2007). These include the network level, organizational level, and community level, and network effectiveness is different for each of them (Provan & Milward, 2001). There may even be tensions between the levels, with organizational goals undermining network goals, or vice versa (Whelan, 2011). For example, what organizations might deem as important for effectiveness may be significantly different from what networks, or the community, deems as important for effectiveness. Network effectiveness also means different things to different networks, dependent on the sector and field in which the network exists (Whelan, 2011). For a network in the public sector providing health or educational services, an effective network might equate to ensuring the affordable provision of services to the public, with community level goals being the most important indicators of effectiveness. On the other hand, a network in the private sector in the field of technology may equate an effective network to one that has a diverse range of products and services, is highly innovative, and is profitable, so organizational and network level goals may be the most
important indicators of effectiveness in the private sector. This research substitutes for the difficulties of measuring objective network effectiveness by measuring network effectiveness subjectively as well. In this sense, we will be looking at perceived trust across the whole network, and effectiveness at the network level. We will not look at effectiveness at the community or organizational levels due to a lack of time, since this would mean including the perspectives of special needs students, their
parents/guardians, teachers, and employers such as the schools and school boards whom are members of the network, to name a few.
To investigate the perceived trust and effectiveness of the network, the following research question will be answered:
How does perceived interorganizational trust within SWV VO 2801 affect the effectiveness of the network as a whole?
In order to answer this research question, the following sub-questions will first be answered: - Is SWV VO 2801 objectively and/or subjectively effective on the network level?
- What types of interorganizational trust are perceived to be present within SWV VO 2801? - How do these types of interorganizational trust affect the effectiveness of SWV VO 2801? The research questions posed above are positive, not normative, since they are explanatory. The focus here is on empirical phenomena, the links between them, and on the links between empirical
phenomena and the theoretical concepts (Toshkov, 2015). Normative research is prescriptive, not explanatory, and is more ethical, dealing with what ought to be instead of what actually is (Toshkov, 2015), which is not the case here.
This thesis and the corresponding research questions posed above are scientifically and societally relevant for several reasons.
Regarding scientific relevance, it is well established that trust facilitates partnerships (Edelenbos & Klijn, 2007), however the specific role that trust plays in facilitating effective networks is largely unknown (Provan & Kenis, 2007; Bunger, 2013; Klijn, Edelenbos, & Steijn, 2010).
Page | 5 Most of the research that has been done in this regard has focused on the private sector mainly
(Lambright, Mischen, & Laramee, 2010), which is outside of the scope of this thesis since we are
focusing on the public sector. The relationship between (perceived) trust and an effective network in the private sector may be different than it is in the public sector because within the private sector, there is usually more competition involved. When multiple organizations rely on the same resources to produce similar products or services, there is competition (Bunger, 2013). The consequence of competition being present within the private sector is that trust, or at least perceived trust, is usually necessary to mitigate its negative effects on coordination between organizations. This is because coordination among
competing organizations is risky. The more organizations coordinate, the more they share control over resources, and the more vulnerable they become to exploitation by their partner(s). When competitors coordinate, the risk of exploitation is even higher because partners have an additional economic interest in each other's failures (Bunger, 2013).
There is a large and growing amount of literature in organizational sociology, contract theory, and business administration on the importance of trust for cooperation and achieving results. In literature on interorganizational cooperation, there is also growing attention to the effects of trust (Edelenbos & Klijn, 2007). Surprisingly then, is the fact that literature on the impact of trust in achieving results in governance networks in public administration, public management, and policy science has been scarce (Klijn, Edelenbos, & Steijn, 2010). If trust is discussed in public management literature, it is mainly focused on citizen trust and not on trust between public actors at the organizational or network levels (Edelenbos & Klijn, 2007; Bunger, 2013).
The few scholars that do discuss trust and networks within the field of public administration usually study trust based on dyadic ties, so it is still largely unknown how whole network trust develops and evolves (Provan, Fish, & Sydow, 2007). Moreover, not much has been studied on the bases of trust and how different trust bases may affect the effectiveness of a network (Poppo, Zhou, & Ruy, 2008). The body of literature on networks is quite extensive and spans across many disciplines (Provan, Fish, & Sydow, 2007; Klijn, Edelenbos, & Steijn, 2010). In general, there has been considerable progress in understanding what networks are, how they are structured, how they operate and are managed, and even how they develop (Agranoff & McGuire, 2001; Klijn, Edelenbos, & Steijn, 2010; Lecy, Mergel, & Schmitz, 2014; Provan, Fish, & Sydow, 2007; Lucidarme, Cardon, & Willem, 2016). Several researchers have even studied whole networks in a public management context, however many of these studies are mainly theoretical (Willem & Lucidarme, 2014; Provan, Fish, Sydow, 2007). Although there are many theoretical studies, empirical studies on the whole network level are lacking. Provan & Milward (2001) suggest a framework to evaluate empirically the effectiveness of networks; however, they did not actually empirically test their framework. Effectiveness at the network level is often discussed, but seldom empirically studied (Lucidarme, Cardon, & Willem, 2016).
Page | 6 There is also a lack of knowledge on mandated networks and how mandates affect network
effectiveness (Provan, Fish, & Sydow, 2007). The role of trust in mandated networks might differ from the role of trust in emergent networks, since in emergent networks; the network, trust, and
commitment develop naturally (Provan & Lemaire, 2012). In mandated networks, organizations may only develop or join networks because they are required to and may be unwilling or reluctant to coordinate with other network partners without trust and commitment. Hence, mandated networks require the building of trust-based relationships and internal legitimacy for them to be effective (Provan & Lemaire, 2012).
Only one piece of literature was found that mentions perceived trust in relation to public networks, and this was mentioned as “an important area for further research,” since perceptions of trust within a network may be affected by a number of different factors (Seppanen, Blomqvist, & Sundqvist, 2007). All of this being said, this thesis is scientifically relevant because it is an empirical study at the whole network level, within the public sector. The type of network being studied is a mandated one. It will attempt to look at the impact of perceived trust on network effectiveness by identifying the different bases of trust present within the network and how they affect network-level effectiveness.
Regarding societal relevance, in practise this thesis may help public managers, public officials, policy makers, and network contact persons to know whether they should include mechanisms to foster trust within a network, as it may allow the network to be more effective. In this case, a more effective network means that within society, secondary schools that provide special education would cooperate as intended within their regional networks, leading to students with special needs receiving the right type of education that suits them.
This thesis will continue in the next section with theoretical arguments on networks, network effectiveness, and trust. Next, the methodological issues of the research will be dealt with in the research design section, which includes an explanation of how the data has been collected and how it will be analysed. The empirical findings and an analysis of these findings will then be presented, with each section answering one of the sub-questions posed earlier. Finally, the thesis will conclude with a summary of its main elements, a concise answer to the research question, suggestions for further research, the strengths and limitations of the research, and an assessment of the validity and reliability of the research.
Page | 7
Theory
Networks
Networks Defined
Provan, Fish, and Sydow (2007) define a network as "a group of three or more organizations connected in ways that facilitate achievement of a common goal." They are multiorganizational arrangements for solving problems that cannot be achieved, or achieved easily, by single organizations (Agranoff & McGuire, 2001). Networks allow for multiple organizations to coordinate to provide otherwise complex, costly, and intractable services, allowing for solutions to societal challenges that would have been difficult, or impossible, for organizations to provide alone (Provan & Lemaire, 2012; Agranoff & McGuire, 2001). They are goal-oriented, are often formally established and governed, and are primarily non-hierarchical, meaning that participants maintain substantial autonomy (Provan & Lemaire, 2012). Although organizational network members maintain substantial autonomy, a public agency may or may not play a lead role in the flow of resources and information among network participants (Provan & Lemaire, 2012).
Sometimes these networks require the coordinated effort of organizations that may not normally coordinate due to diverging organizational goals and operational routines (Lundin, 2007). Thus, if a network is to be truly effective, it must be able to operate through cooperation and collaboration fostered by trust and commitment between the organizations within it. Therefore, all networks involve a mix of both contractual and more informal, trust-based ties that complement each other (Provan & Lemaire, 2012).
Many public challenges that face modern societies, such as health and social needs, unemployment and workforce development, community and regional economic development, and disaster preparedness and emergency response, are difficult to tackle using just one single public authority (Lundin, 2007). In order to address these challenges, networks have become increasingly popular (Provan & Lemaire, 2012). This is because there is a diverse range of resources and knowledge spread amongst multiple organizations, and the coordination of these organizations within networks allows for increased flexibility/adaptability, enhanced learning, improved knowledge flow, and greater sensitivity to the needs of clients, which are all needed when solving the problems that modern societies face (Provan & Lemaire, 2012).
Types of Networks
Networks can be mandated or emergent. Each of these types of networks have their own costs and benefits, however, whether a network is mandated or emergent, for it to be effective and to function the way it was intended to, it must be able to sustain itself through trust and commitment (Provan & Lemaire, 2012). Agranoff and McGuire (2001) take this a step further by emphasizing the importance of trust, common purpose, mutual dependency, resource availability, catalytic actors, and managerial ability within all networks.
Page | 8
Mandated Networks
In a mandated network, the push for collaboration among organizations comes from government, and government agencies play a much more significant role in initiating the formation of and sustaining these networks than they do in emergent networks (Provan & Lemaire, 2012). With a top-down
mandate to form or join a network, organizations have powerful incentives, such as funding, to attempt to work together. Because of this, this form of network is appropriate in situations where a network is necessary and where said network would not develop, or would develop slowly, without incentives. This situation may occur when the organizations within the network have no other reason to work together other than the mandate from government, or when organizations exist in a competitive environment (Provan & Lemaire, 2012). Because of the external mandate to network, external legitimacy is usually easy to establish for these networks, while there is a risk that internal legitimacy is ignored or
undervalued. Some reasons that mandated networks may not operate as intended are this lack of internal legitimacy stemming from a lack of homophily, friendship, trust, or the need to acquire legitimacy or power (Provan & Lemaire, 2012).
Emergent Networks
On the other hand, emergent, bottom-up networks develop on their own, with no external push or incentives. These networks do not have issues of building trust and commitment because this is able to occur naturally (Provan & Lemaire, 2012). Trust here is related to obligations and expectations, and these assumed obligations and expectations are essential in holding networks together (Agranoff & McGuire, 2001). In fact, in an emergent network, internal legitimacy is usually strong because of the presence of trust and commitment (Provan & Lemaire, 2012). The disadvantages of these networks are that they are likely to have difficulties in externally legitimizing the network to new or potential
members. Aside from external legitimacy, emergent networks also have problems with sustaining themselves on their own, over processing, and network governance. This is because they require significant time and effort to be spent on coordinating network activities, and with no hierarchy, establishing a network governance form that allows everyone to keep their autonomy is difficult, even when trust and commitment are present (Provan & Lemaire, 2012). Some of the main factors that explain the emergence of emergent networks include homophily, proximity, heterophily, the need to reduce dependence on others, prior relationship experience, the need to gain legitimacy, and access to key information and/or resources (Provan & Lemaire, 2012).
Network Effectiveness
Network Effectiveness Defined
Network effectiveness is defined as the attainment of positive network level outcomes that could not normally be achieved by individual organizational participants acting independently (Provan & Kenis, 2007). It is hard to provide an exact example of network effectiveness, since this will mean different things to each network and to each sector in which a network exists (Whelan, 2011). That being said, network effectiveness includes the effects, outcomes, impact, and benefits that are produced by a network. It can lead to increased efficiency, client satisfaction, increased legitimacy, resource acquisition, and reduced costs (Lucidarme, Cardon, & Willem, 2016).
Network effectiveness may be viewed objectively; in terms of network stability and financial status, and subjectively; through the perceptions of stakeholders and differing criterion such as outcome, input, and process (van Raaij, 2006).
Page | 9 Measuring network effectiveness objectively means determining whether a particular network is
meeting its objectives (Whelan, 2011). This can be beneficial in order to determine the economic position of the network and/or the benefits of network participation (van Raaij, 2006).
Subjective network effectiveness includes the perceptions of stakeholders and network members and the criteria they use such as outcome, input, and process criteria (van Raaij, 2006). It is beneficial to understand the norms and criteria used by network members since networks are usually at least partly self-governed. Members have their own criteria, which drives their behaviour within the network (van Raaij, 2006). Outcome criteria could include enhanced client well-being, goal accomplishment, and stakeholder satisfaction. Input criteria could include sufficient resources, while process criteria could include the absences of disturbing relations within the network and between network members and clients (van Raaij, 2006).
Levels of Network Effectiveness
Network effectiveness can be viewed from three different levels: the network itself, the network’s (organizational) members, and the community the network serves (Provan & Milward, 2001). In practise, all three of these levels are of equal importance when discussing the effectiveness of a network, especially for a network in the public sector. This is because a public-sector network uses public resources and delivers services to the public, so it should be effective on all levels and in all regards (Lucidarme, Cardon, & Willem, 2016). Despite all that, this thesis focuses solely on the network level of effectiveness, for the reasons discussed in the Introduction.
The Network Level
The simplest way to evaluate network-level effectiveness in emergent networks is the rate of network membership growth, which differs at different stages of maturity of the network (Provan & Milward, 2001). This is because during the early stages of an emergent network it needs to be able to attract and retain members, which might be especially difficult if the network has not been able to establish external legitimacy yet (Provan & Lemaire, 2012). However, once a network is established then attracting more organizations becomes less important. The reason for this is that after surpassing a certain size, networks will begin to become less effective due to increasing coordination costs (Provan & Milward, 2001), especially emergent networks, which often have difficulties with over processing and sustaining themselves (Provan & Lemaire, 2012).
Another way of assessing network-level effectiveness is by the range of services provided by the network (Provan & Milward, 2001). Networks allow for the provision of a broad range of services by multiple organizations that address all the needs of their clients, however, if this is not the case in practice, then it can be said that the network is less effective (Provan & Milward, 2001). On one hand, a limited range of services may be offered by the organizations within a network, leaving some clients unsatisfied. On the other hand, networks may end up providing too many overlapping or duplicate services, resulting in confusion (Provan & Milward, 2001). For these reasons, network-level effectiveness should be judged partly by the extent to which only those services that are actually needed are being provided (Provan & Milward, 2001).
Page | 10 A third way to evaluate network-level effectiveness is to assess the strength of the relationships
between and among network members in dyads as well as across the full network (Provan & Milward, 2001). In the early stages of the network, ties may be uncertain, calculated, and weak, however over time these may become stronger, especially when there is high multiplexity (Provan & Milward, 2001). Finally, network-level effectiveness may be assessed by looking at member commitment to network goals (Provan & Milward, 2001). This is because networks depend exclusively on the involvement and commitment of all, or at least a significant subset, of the organizations that comprise the network (Provan & Kenis, 2007). Without commitment to network goals, network participants are less likely to be involved or work together, which undermines the entire idea of networks in the first place. In the worst cases, a lack of commitment to network goals would lead to the network existing in name only (Provan & Kenis, 2007).
The Organizational Level
One can assess effectiveness at the organization level in numerous ways, such as client outcomes, agency survival, enhanced legitimacy, resource acquisition, cost of services, service access, and
minimum conflict for multiprogram agencies across multiple networks (Provan & Milward, 2001). This is important since individual organizations and their managers are partially motivated by self-interest, and so they will seek the benefits of joining a network before they do so (Provan & Milward, 2001).
The Community Level
Most problems in the public sphere are community problems that need to be addressed at the
community level. Community-based networks need to be judged based on the contribution they make to the communities they serve in (Provan & Milward, 2001).
There are different ways to measure the community-level effectiveness of a network such as the overall costs of service provision, the extent to which the network has assisted in building social capital, public perceptions that the social problem is being solved, changes in the incidence of the problem, and aggregate indicators of client well-being (Provan & Milward, 2001).
Benefits of Network Effectiveness
Effective networks bring about institutional isomorphism or imitation through information exchange and learning (DiMaggio & Powell, 1983; Willem & Lucidarme, 2014). Organizations are more likely to imitate processes if they are tied to other organizations whom have successfully employed those processes. Trust-based network ties influence organizations to evolve through institutional isomorphism (DiMaggio & Powell, 1983). Organizations within a network may learn from one another and evolve in ways that lead to network effectiveness (Provan, Fish, & Sydow, 2007).
Effective networks are sustainable, bring about enhanced information processing capabilities, and have the capacity to address the needs of both their organizational members and other key stakeholders (Provan & Lemaire, 2012).
More benefits of effective networks include innovation, survival of the organizations in the network, enhanced performance of the organizations in the network, and economic benefits (Willem &
Lucidarme, 2014; Provan & Lemaire, 2012). These benefits of network effectiveness obviously stem from trust (Willem & Lucidarme, 2014; Agranoff & McGuire, 2001).
Page | 11
Trust
Trust Defined
While there is no one universally accepted definition of trust, there is a consensus that in order for trust to exist, there must be some risk and interdependence involved (Lambright, Mischen, & Laramee, 2010). According to Edelenbos & Klijn (2007), the three most dominant characteristics of trust are vulnerability, risk, and expectation. Trust can thus be defined as an aspect of a relationship that reflects the
willingness to accept vulnerability, or risks, based on positive expectations about another's intentions or behaviours, or interdependence (Provan & Kenis, 2007).
When trust exists, there is the expectation that an organizational partner will not behave
opportunistically or exploit a partner's vulnerability, instead demonstrating fairness, reliability, and goodwill (Bunger, 2013; Klijn, Edelenbos, & Steijn, 2010). If one actor assumes good intentions on the part of the other, the likelihood of any unexpected interactions because of opportunistic behaviour is lower (Klijn, Edelenbos, & Steijn, 2010). Thus, when there is trust, one actor believes and expects that the other actor will take both actors’ interests into account (Klijn, Edelenbos, & Steijn, 2010), and that power will not be used or abused at the cost of others in the network (Willem & Lucidarme, 2014). Combined, this definition and these descriptions form the basis for the types of trust that are referred to throughout this thesis.
Types of Trust
Now that we have defined trust, we can get into the distinctions between the different types of trust, or trust bases, which were found in the literature. Table 1 helps to visualize the spectrum of trust bases listed and explained below.
Table 1. The Different Bases of Trust. Amended from Market, Hierarchy, and Trust: The Knowledge Economy and the Future of Capitalism (p. 218), by P. S. Adler, 2001, Organizational Science.
Dimensions Components Bases - Consistency/contractual - Competence/cognition - Benevolence/loyalty/concern/affect/goodwill - Honesty/integrity - Openness
These bases include consistency-based trust, competence-based trust, benevolence-based trust,
honesty-based trust, and openness-based trust (Adler, 2001; Willem & Lucidarme, 2014). These bases of trust build upon one another, are complementary, and can be placed on a spectrum. This means that, in theory, openness would lead to honesty, which would lead to benevolence, then competence, and finally consistency, the strongest base of them all (Adler, 2001).
Consistency-based Trust
Consistency-based trust is also referred to as contractual trust. This form of trust is based on keeping promises. According to Sako (1992), any transaction between organizations relies on this form of trust for successful execution. This form of trust is reflected in both written and oral agreements (Sako, 1992).
Page | 12
Competence-based Trust
Competence-based trust, or cognition-based trust, refers to trust based on rational reasons and evidence (Willem & Lucidarme, 2014). This trust base is concerned with the expectation of
organizational partners performing their tasks competently and to a certain professional standard. The higher competence-based trust is, the less need there is for inspections and quality assurance amongst organizations within a network (Sako, 1992).
Benevolence-based Trust
Benevolence-based trust stems from loyalty, concern, or goodwill. It is sometimes referred to as affect-based trust. It refers to trust affect-based on emotions and beliefs that others will help care and maintain a good relationship (Willem & Lucidarme, 2014). Trust based on benevolence involves mutual
expectations of open commitment to one another, where commitment is defined as the willingness to do more than what is formally expected (Sako, 1992). This means that there are no explicit agreements to be fulfilled, such as with consistency-based trust, and no professional standards to be met, which is the case with competence-based trust. Goodwill-based trust allows for dependability and a high level of discretion (Sako, 1992).
Honesty-based Trust
Next, honesty-based trust, also referred to as integrity-based trust, is rooted in perceptions about an organization's motives and character and emphasizes the social and attitudinal underpinnings of the relationship (Connelly, Crook, Combs, Ketchen, & Aguinis, 2018).
Openness-based Trust
Finally, we have trust based on openness. According to Chou & Chiang (2013), openness-based trust is based on an organization being completely open and dedicated and involves all the prior trust bases, including consistency and competence, but most importantly, benevolence and honesty. This is because this trust base stems from organizations having both goodwill and honest communication (Chou & Chiang, 2013).
Trust in Networks
Trust is frequently discussed in general network literature as critical for network performance and sustainability (Provan & Kenis, 2007; Provan & Lemaire, 2012; Agranoff & McGuire, 2001; Lambright, Mischen, & Laramee, 2010; Bunger 2013).
Networks have no hierarchical structure, making them a unique governance method when compared to the two main types of governance, markets and hierarchies (Willem & Lucidarme, 2014). This lack of hierarchy leads to uncertainty, risks, and costs such as concerns of opportunism, threats to autonomy, and the costs of monitoring partnerships (Lambright, Mischen, & Laramee, 2010). The lack of a hierarchical structure also makes it so that there are fewer mechanisms in place within a network to ensure lasting, successful operations than when compared to hierarchies and markets, making trust necessary to replace these mechanisms. Trust is related to obligations and expectations, and these assumed obligations and expectations are essential in holding networks together (Agranoff & McGuire, 2001). The existence of reciprocal relationships based on trust is a necessary condition for the existence of networks (Willem & Lucidarme, 2014).
Page | 13 At the same time, one must take in to consideration that networks might exist and survive even with low levels of trust among their members. This is especially the case in networks where public, profit, and non-profit are mixed; trust is not obvious and might be low or weak (Willem & Lucidarme, 2014). Weak trust is the expectation that communicated preferences are honest instead of being misleading, and that a commitment will stick if the conditions under which it was entered have not drastically changed (Lundin, 2007). When there is at least weak trust, risks can be taken based on the belief that the other party can be trusted. If there is not even so much as weak trust, the fear is that actors will refrain from acting and cooperating (Edelenbos & Klijn, 2007). Goal-oriented public networks that bring together organizations do not initially have trust as their main governance mechanism, such as in mandated networks. Because network structure is quite important, sometimes trust is not truly necessary if the network structure exists (Provan & Lemaire, 2012; Willem & Lucidarme, 2014).
Previously, a distinction was made between several bases of trust (Adler, 2001). Out of the bases of trust that were identified, competence- and honesty-based trust were deemed especially important by scholars (Adler, 2001; Edelenbos & Klijn, 2007; Willem & Lucidarme, 2014). This is because in the past, traditional forms of trust were based mostly on loyalty. Over time, trust has become more rational, reflective, inclusive, and open (Adler, 2001). Things such as competence and honesty are now deemed more important than loyalty (Willem & Lucidarme, 2014), since the expected gains of cooperation are important (Edelenbos & Klijn, 2007). Empirically, all the other bases of trust have proven to be complementary, since they build upon one another, and hence are all important for network effectiveness (Adler, 2001).
Edelenbos & Klijn (2007) carried out a case study that showed that only after recognizing the
competences of organization B did organization A begin to place trust in organization B. According to them, at was at that moment that organization A became more willing to cooperate with organization B and trust them with their business (Edelenbos & Klijn, 2007). They distinguish two kinds of competence-based trust, namely (a) functional or specific competence: competence in the specified knowledge and skills required to do a particular job; and (b) interpersonal competence: people skills and knowing how to work with people (Edelenbos & Klijn, 2007). Other scholars studied by Willem & Lucidarme (2014) also indicate the importance of competence-based trust for similar reasons.
Benefits of Trust
When discussing network effectiveness, trust plays a key role. The absence of trust within a network may serve as an important explanation for its ineffectiveness (Lecy, Mergel, & Schmitz, 2014).
Trust holds networks together by facilitating inter-organizational cooperation, solidifying cooperation, and enhancing network performance (Edelenbos & Klijn, 2007). In the remainder of this section, these three benefits of trust in networks are discussed.
Page | 14 First, trust facilitates inter-organizational cooperation by making it both possible and cheaper
(Edelenbos & Klijn, 2007). If organizations do not trust each other, resource interdependence and goal congruence do not matter because they will not cooperate (Lundin, 2007). Therefore, trust within a network directly influences how closely organizations coordinate their administrative operations. A lack of trust would lead to organizations being hesitant to align operations, or to them doing so only limitedly (Bunger, 2013). Trust counteracts this hesitation since it has a positive impact on expected future
cooperation within a network (Lambright, Mischen, & Laramee, 2010). Administrative coordination is viewed as less risky, leading to organizations within the network sharing greater amounts of
administrative resources and coordinating more with one another (Bunger, 2013). Trust serves to reduce the costs of complex cooperation processes since there is less need for details and specifications within these processes when trust is present (Klijn, Edelenbos, & Steijn, 2010).
Next, trust solidifies cooperation improving robustness and investments in cooperation (Edelenbos & Klijn, 2007). As organizations within a mandated network interact and become familiar with one another, trust builds through their interaction (Bunger, 2013), leading to higher levels of investment in the network. If something goes wrong within a network, partners that trust one another can handle this more easily based on a willingness to believe that it is no one's fault and that everyone within the network did their best to comply with the rules and agreements made (Edelenbos & Klijn, 2007). This goes hand-in hand with the fact that trust reduces the risks of uncertainty inherent to networks because it creates greater predictability (Klijn, Edelenbos, & Steijn, 2010; Seppanen, Blomqvist, & Sundqvist, 2007). This directly relates to the theory that when there is more trust, there is less need to monitor regularly network compliance, since future cooperation and compliance are assumed in the presence of trust (Agranoff & McGuire, 2001; Seppanen, Blomqvist, & Sundqvist, 2007). Bunger (2013) makes a similar claim that trustworthiness reduces the costs of monitoring partnerships, resulting in organizations spending less time on developing formal agreements and monitoring compliance. Finally, trust enhances network performance by providing organizations within a network with the capacity to stretch and leverage limited resources and increase their problem-solving capacity (Provan & Lemaire, 2012; Edelenbos & Klijn, 2007). The presence of trust in a network stimulates learning and the exchange of information and knowledge within the network (Klijn, Edelenbos, & Steijn, 2010; Seppanen, Blomqvist, & Sundqvist, 2007). Some information is specialist and is not always easy to trade, resting in the expertise of persons and/or organizations (Edelenbos & Klijn, 2007). The involvement of these different stakeholders in networks generates more information and knowledge (Klijn, Edelenbos, & Steijn, 2010; Seppanen, Blomqvist, & Sundqvist, 2007). With knowledge only being available in the form of human capital, it can only be acquired through exchange, cooperation, and learning. This information and knowledge can then be used throughout the network to establish better-tailored solutions to the societal problem the network was created to tackle (Klijn, Edelenbos, & Steijn, 2010). In situations where trust is present, the flow of information and the willingness to exchange information is likely to be greater amongst organizations, resulting in a larger problem-solving capacity (Edelenbos & Klijn, 2007).
Page | 15 The multiorganizational aspect of networks combined with trust amongst these organizations allows for enhanced learning, which leads to greater sharing of information; innovation and better service quality; the provision of an enhanced range of services to clients who no longer have to deal separately with different providers; the ability to achieve economies of scale and greater competitiveness; the ability to exert more pressure on politicians and funders due to increased influence and reach; the capacity to be more flexible than traditional bureaucracies; and enhanced capacity and responsiveness to deal with unforeseen circumstances (Provan & Lemaire, 2012).
Now that the theoretical arguments to be used in this thesis have all been discussed, the following section will explain the research design that was employed.
Page | 16
Research Design
The research undertaken in this thesis is explanatory and positive in nature because it attempts to explain how different types of perceived interorganizational trust affect network-level effectiveness. It does so by first determining whether the network in question, SWV VO 2801, is effective or not on the network level. Once this has been determined, the thesis will continue by determining what types of perceived interorganizational trust are present within SWV VO 2801, and finally, by drawing conclusions on whether and how the different types of perceived interorganizational trust found within SWV VO 2801 affect its network-level effectiveness.
This means that a retrospective approach was taken for this study, where the outcome – a network effective on at least the network level – was already known. The focus here was on finding out whether perceived interorganizational trust contributed to that or not, and how.
This research may be described as empirical because it draws from real-life phenomena that occur within the network in question, SWV VO 2801, and empirical facts and observations about the network collected during interviews with network members.
The research is said to be theoretically informed because it uses theory in an attempt to describe and explain what is happening in reality. For example, theories on interorganizational trust were used to determine the types of interorganizational trust perceived to be present within the network. The objective of this thesis is to test the theories discussed earlier, in the Theory section, in order to either confirm or refute them. Theory testing can be accomplished using a variety of research methods, however, for the sake of this research; a within-case study will be carried out. Using a single-within-case study to test theory has its limitations – mainly which the conclusions drawn from this research are not generalizable at all and are relevant to this one specific case. Still, this thesis remains scientifically and socially relevant for the many reasons discussed in the Introduction.
The single-within-case study has been carried out in the form of an in-depth study of one of the 76 regional secondary school networks in the Netherlands. This research type was chosen since it allows for the collection of dense, rich qualitative data, such as perceptions of interorganizational trust within the network, and how these affect the network’s effectiveness (on the network level).
Case Selection
The population being studied for this research is that of regional education networks in the Netherlands. The case is taken from one of these regional networks, Samenwerkingsverband Passend Onderwijs VO 2801. As stated in the Introduction of this thesis, the network consists of 12 schoolboards with 26 secondary schools located in the regions of Voorschoten, Wassenaar, Leiderdorp, Zoeterwoude, Oegstgeest, Leiden, and Kaag & Braassem.
The case was chosen based on convenience sampling. The researcher regarded this network as accessible due to the close proximity of the regions in question, since the researcher is located in Den Haag. The researcher did not need to exhaust many resources when interviews needed to be conducted, such as travel time and transportation fees. This made SWV VO 2801 more feasible than other networks that fall under the same policy, since those would call for more travelling and hence, resources.
Page | 17 The director of SWV VO 2801 is:
- Annemarije van Overschot
The coordinator of the Loket Passend Onderwijs SWV VO 2801 is: - Erik Beers
The organizational members of SWV VO 2801 are the following schoolboards, all of which have at least one school location within the regions listed previously within this section:
- Stichting Het Rijnlands Lyceum - Stichting Wellant
- Stichting Confessioneel Onderwijs Leiden - Stichting Scholengroep Leonardo da Vinci Leiden - Stichting Praktijkonderwijs Leiden e.o.
- Stichting Stedelijk Gymnasium Leiden - Stichting Vrije Scholen Zuidwest Nederland - Stichting S. Adelbert College
- Stichting voor Voorgezet Onderwijs op Interconfessionele Grondslag - Stichting Speciaal Onderwijs Leiden
- Stichting Prof. Dr. Leo Kanner Onderwijsgroep - Stichting Fioretti Teylingen
Operationalisation
In this case, the key concepts are network effectiveness on the network level, and perceptions of interorganizational trust within the network. Network effectiveness is the dependent variable, while the perceptions of different types of interorganizational trust within the network serve as the independent variables.
The concept of network effectiveness is operationalized in Table 2. As shown in the table, the concept of network effectiveness has been operationalized to include both the objective effectiveness of the network and the subjective effectiveness of the network.
Objective network effectiveness was measured using the following operational indicators: - Financial status
- Range of services provided by the network - Goal accomplishment
These operational indicators originate from the theories on network-level effectiveness discussed in the Theory chapter of this thesis. They have been amended to fit this specific case, so indicators that are relevant to emergent networks have been left out, since we are dealing with a mandated network.
Page | 18
Table 2. Operationalization of Network Effectiveness.
Concept Variables Operational Indicators Source of Information Network
Effectiveness
Objective Network Effectiveness
Financial status Documents -
Inspectorate of Education Report & Financial Statement 2017 Range of services provided by
the network
Document -
Network Support Plan Goal accomplishment Statistics of dropouts in secondary education Website - VSV in de Gemeenten: Onderwijs in Cijfers Subjective Network Effectiveness
Stakeholder satisfaction Interviews Member commitment Interviews Strength of relationships Interviews
In the theory, goal accomplishment was labelled as a subjective indicator, and member commitment to network goals and the strength of relationships between and among members across the whole network were labelled as objective indicators. These have also been amended, since the researcher finds that member commitment and the strength of relationships is quite subjective, and that goal accomplishment is more objective. The sources of information for these indicators, as shown in Table 2, also contribute to their new labels, since documentation is more objective, and interviews are inherently subjective.
The sources of information for the objective measures of network effectiveness were all different forms of documentation.
The financial status of the network was operationalized using information found within a report published by the Inspectorate of Education here in the Netherlands, titled "Stichting
Samenwerkingsverband Passend Onderwijs VO 2801: Onderzoek bestuur en samenwerkingsverband", which is translated in English to: Partnership Inclusive Education VO 2801: Research on the board and partnership. The data in this report comes from the network support plan and annual reports published by SWV VO 2801. The Inspectorate also requested access to several additional documents from the network, and a series of round-table meetings were held with a group of parents, teachers, school principals, and professionals working within youth care, that also included the Loket Passend Onderwijs SWV VO 2801, the board of the network, and the director of the network.
The financial status of the network was also operationalized using information from SWV VO 2801’s yearly financial statements, specifically that of the year 2017. This document, entitled “Stichting Samenwerkingsverband Passend Onderwijs VO 2801: Jaarrekening 2017”, and several similar reports and publications, were found on the website of the SWV.
Page | 19 The range of services provided by the network was operationalized using a different document, namely the “Ondersteuningsplan Samenwerkingsverband Passend Onderwijs VO 2801”, translated in English to: Support Plan Partnership Inclusive Education VO 2801. This support plan covers the years 2016-2020. Every four years, each regional network is required to publish their own support plan which explains how the network tackles its mission of ensuring that all students in secondary education within its region receives a form of education that suits their development opportunities.
Goal accomplishment was operationalized by using national statistics on the number of new dropouts per schoolyear since the policy was introduced in 2014. This indicator was chosen since it was also used by interview respondents when asked whether they found SWV VO 2801 to be effective. The thought process behind this is that if students are not receiving a suitable form of education, they will be forced to drop out of school. The information was documented on a website, namely “Onderwijs in Cijfers: Voortijdig Schoolverlaters (VSV) in de Gemeenten”, which literally translates in English to Education in Figures: Early School Leavers in the Municipalities. The information on this website is consigned by the Ministry of Education, Culture, and Science in the Netherlands, DUO (Dienst Uitvoering Onderwijs), and CBS (Centraal Bureau voor de Statistiek). The information is supplemented by the UWV
(Uitvoeringsinstituut Werknemersverzekeringen) and the register of exemptions for the Compulsory Education Act.
Subjective network effectiveness was operationalized by using the perceptions of the network
coordinator and several network members that were collected during interviews, as shown in Table 2. These criteria and the different corresponding dimensions used to operationalize their perspectives are as follows:
- Outcome criteria:
o Stakeholder satisfaction - Process criterion:
o Member commitment to network goals
o Strength of relationships between and among network members
The concept of perceptions of interorganizational trust is defined here as how network stakeholders perceive interorganizational trust within their network. Specifically, the perceptions of the network coordinator and several network members collected during interviews will be used in order to
determine the different types of interorganizational trust perceived to be present within SWV VO 2801. These include any of the following types, or bases, of interorganizational trust:
- Consistency-based trust - Competence-based trust - Benevolence-based trust - Honesty-based trust - Openness-based trust
After determining whether the network is effective on the network level, either objectively, subjectively, or both, and the different types of trust perceived to be present within the network have been
identified, theory-informed conclusions will be drawn. These conclusions explain how the different types of perceived trust affect network-level effectiveness within SWV VO 2801.
Page | 20
Method of Data Collection
In order to collect the data needed to answer the research question posed in the Introduction of this thesis, desk research was used to find the secondary documents that were introduced in previous Operationalization section. Secondly, field research was done in the form of interviews, which were conducted in order to collect subjective data, such as subjective network effectiveness and perceptions of trust within SWV VO 2801.
The interviews were semi-structured. This meant that all the respondents in a sub-group were asked the same questions, while also allowing them to give any additional information they deemed necessary or important. This structure of interview ensured that the researcher would be able to collect all the data necessary and infer the results of the data collected without sacrificing the validity and reliability of the interviews.
The first interview respondent was the coordinator of the Loket Passend Onderwijs SWV VO 2801, Erik Beers. For the sake of clarity, the ‘Loket Passend Onderwijs SWV VO 2801’ will be referred to as ‘the Loket’ throughout the remainder of this thesis. Mr. Beers is referred to simply as the network
coordinator or the coordinator of the Loket. The preference to interview the coordinator of the Loket instead of the network director was based on two things. First, someone in his position is more hands-on and active within the network and amhands-ongst school locatihands-on directors, and hence, might be more knowledgeable of interorganizational trust relations amongst them. Because of this, it was assumed that his perception of interorganizational trust within the network would be more accurate than the network director’s perception. Secondly, the current network director, Ms. Annemarije van Overschot, was only recently appointed, in March 2018, and hence might not have had the experience necessary to answer some of the questions that were asked. The interview report can be found in Appendix A.
The second sub-group of interview respondents consisted of members of all 12 schoolboards that make up the network. Although interview requests were sent to every schoolboard, interviews were not conducted with all. Here follows a list of the five schoolboard members that were interviewed, along with information on their position within the SWV VO 2801 and the schoolboards that they represent:
- C. H. P. Vreugdenhill, Member of the General Board SWV VO 2801, Stichting Fioretti Teylingen
- Dr. M. W. Knoester, Member of the Executive Board SWV VO 2801, Stichting Het Rijnlands Lyceum
- H. Freitag, Member of the Executive Board & Treasurer SWV VO 2801, Stichting Scholengroep Leonardo da Vinci Leiden
- R. Timmer, Member of the Executive Board & Rep. for Special Education SWV VO 2801, Stichting Prof. Dr. Leo Kanner Onderwijsgroep
- F. W. Hoekstra, Chairman of the Executive Board SWV VO 2801, Stichting Confessioneel Onderwijs Leiden
All the interviews were conducted in person except for one, namely that of Dr. Knoester. Conducting the interviews in person meant that the researcher could observe any notable body language, which may have indicated biased or dishonest responses. This was not the case.
Page | 21 Because interviews were used to obtain some of the data needed to answer the research question, informed consent was necessary from every respondent. In order to obtain informed consent, the researcher ensured that the respondent was aware that participation as a respondent was voluntary. The researcher also ensured that the interviewee was aware of the purpose of the research, that the respondent gave permission to the researcher to take notes and record the interview, and that the respondent knew that he had the right not to answer questions and to withdraw from the interview at any time. Finally, the researcher guaranteed confidentiality if it was requested, however this request was not made by any of the respondents. The interview reports may be found in Appendix B.
Method of Analysis
Once the data was collected, deductive reasoning was used to analyse the data and answer the research question. When using the deductive reasoning approach, there is already a predetermined framework of theories that the researcher uses the collected data to test.
Coding has been chosen to allow the researcher to organize and analyse the dense qualitative information collected during the interviews in a more structured, reliable way.
Staying true to the deductive approach taken throughout this thesis, a priori coding was used to
determine the first level of codes that were used to analyse the data collected from the interviews. This means that pre-existing theoretical frameworks were used to determine the codes used in the initial phase of coding. These pre-existing theoretical frameworks stem from the theories on subjective network effectiveness and the different bases of interorganizational trust that were discussed in the Theory chapter of this thesis. Furthermore, both in vivo and descriptive coding were used to determine the second, more in depth level of codes. The application of these different forms of coding could be referred to as simultaneous coding.
Because of their lengthiness, the codebooks for both subjective network effectiveness and perceptions of interorganizational trust may be found in Appendix C. The codebooks include the codes themselves, descriptions of the codes, and examples for each.
Page | 22
Empirical Findings & Analysis
Now that the thesis has been introduced, the theory has been discussed, and the research design has been presented, it is possible to present and analyse the data collected in order to answer the main research question and sub-questions presented in the Introduction. In this section, the sub-questions will first be answered. Once this is done, the information gathered to answer the sub-questions will be used to answer the main research question.
Network Effectiveness
Here, we discuss the first research question, which is as follows:
- Is SWV VO 2801 objectively and/or subjectively effective on the network level?
This sub-question will be answered by first analysing and presenting the objective measures of network effectiveness on the network level. Next, the subjective measures of network effectiveness on the network level will be analysed and presented. Finally, the results of both of these will be combined in order to determine the overall network effectiveness of SWV VO 2801 on the network level.
Objective Network Effectiveness
The objective network effectiveness of SWV VO 2801 has be determined using three operational indicators, namely (1) financial status, (2) range of services provided by the network, and (3) goal accomplishment. As indicated previously in Table 2, located in the Research Design, the sources of information for these indicators were the report written by the Dutch Inspectorate of Education, a financial report published on the network’s website for the year 2017, the network support plan, and statistics on dropouts within secondary education across municipalities here in the Netherlands, respectively. These sources of information were used since they are objective in nature and verifiable.
Financial Status
The financial status of SWV 2801 is healthy, based on information collected both from the inspection report and the network’s financial statement from 2017 (Inspectie van het Onderwijs, 2018; SWV VO 2801, 2017). The Inspectorate's feedback on the financial status of SWV VO 2801 is based on the
underlying financial standards of continuity and regularity, which were both assessed as sufficient within the report (Inspectie van het Onderwijs, 2018).
Shareholder's equity at the end of 2016 amounted to €3,982,861 (Inspectie van het Onderwijs, 2018). The network managed to maintain its financial status the following year, with the result of operations by the end of 2017 amounting to a surplus of €420,105. This means that by the end of that year,
shareholder’s equity had increased to €4,402,966 (SWV VO 2801, 2017). With an estimate of €1 million needed to cover any risks, there would still be almost €3-4 million remaining in shareholder’s equity. This means that both short- and medium-term financial continuity in this network are guaranteed (Inspectie van het Onderwijs, 2018).
Financial regularity, or legality, has been assessed as sufficient because of the use of external financial auditors and other investigations made into the network’s finances. There was nothing found to indicate that there were any irregularities (Inspectie van het Onderwijs, 2018).
Page | 23 The healthy financial status of SWV VO 2801 has a positive effect on the network's effectiveness, as confirmed by the theory. According to Agranoff and McGuire (2001), resource availability is important for all networks, not only in the private sector, where competition between organizations plays a large role (Bunger, 2013), or in emergent networks, where access to resources is a driving force for the establishment of a network (Provan & Lemaire, 2012).
Range of Services Provided by the Network
According to the network support plan, each school within the network is required to publish a school support profile that describes the different forms of support that school can offer. All the schools in the network are required provide to at least the basic support that is agreed upon on the network support plan, in addition to whatever extra forms of support the school chooses to offer (SWV VO 2801, 2016). The schools in the network offer two forms of basic support, namely prepatory secondary vocational education (LWOO) and labour market-oriented education (PrO), and one form of special support, special secondary education (VSO). There is also an international bridging class offered within the network for students with difficulties mastering the Dutch language (SWV VO 2801, 2016). In addition to these mainstream forms of education, when a school cannot offer a student the support he or she needs, arrangements can be requested from the Loket, and if they are approved, provisions are made. These arrangements may be made both within schools and between schools (SWV VO 2801, 2016).
For these reasons, it is safe to say that SWV VO 2801 provides a range of services that is necessary for its consumers, or students in this case, and that whatever is not provided can be requested and arranged. The range of services provided by SWV VO 2801 affects the network's effectiveness in a positive way. According to the theory, networks may provide either too little or too many services, leaving their clients either unsatisfied or confused and overwhelmed (Provan & Milward, 2001). SWV VO 2801 offers a basic range of educational services, but also allows additional arrangements to be made for those students whom are left unsatisfied with the more traditional forms of education offered within the network. These arrangements need to be requested and approved before they are made, helping the network to avoid the provision of too many overlapping or unnecessary services. This means that no student is left unsatisfied, but there are also no chances for him or her to be confused and overwhelmed with the number of services provided.
Goal Accomplishment
Because the policy for suitable education went into place in 2014, national statistics of secondary school dropouts from that school year onwards have been used in order to see whether there was any
difference in these numbers since the onset of the policy and the formation of the networks. These statistics include students from both regular and special secondary schools within the municipalities indicated. A comparison between dropout rates before and after the introduction of the policy might have also been a good way to see whether there were any significant changes after the policy was introduced, however statistical information could not be found for that timeframe.
As shown in Figure 1, in 2015/2016, when compared to the dropout percentages of the previous year, 2014/2015, four out of the seven municipalities that fall under SWV VO 2801 recorded a decrease in dropout percentages. These were Kaag & Braasem (-0.12%), Leiden (-0.60%), Leiderdorp (-0.36%), and Oestgeest (-0.15%). Voorschoten maintained the same percentage of dropouts during that year, while Wassenaar (+0.29%) and Zoeterwoude (+0.32%) saw an increase in the percentage.
Page | 24 When comparing the year 2016/2017 to the year 2015/2016, only three of the seven municipalities saw a decrease in dropout percentages, while the remaining four municipalities experienced an increase in the number of dropouts. Oestgeest, Voorschoten, and Zoeterwoude managed to decrease their numbers of dropouts by -0.10%, -0.34%, and -0.58% respectively that year, while Kaag & Braasem, Leiden, Leiderdorp, and Wassenaar experienced increases of +0.51%, +0.36%, +0.56%, and +0.08% respectively.
On average, four of the seven municipalities that fall within the region of SWV VO 2801 experienced an overall decrease in dropouts at the secondary level of education from 2014/2015 through 2016/2017, as seen in Figure 1. These include Leiden (-0.24%), Oestgeest (-0.25%), Voorschoten (-0.34%), and
Zoeterwoude (-0.26%). The three remaining municipalities recorded overall increases in dropouts, specifically Kaag & Braasem (+0.39%), Leiderdorp (+0.20%), and Wassenaar (+0.37%).
Only Oestgeest managed to maintain the downward trend of dropouts over the three-year timeframe, experiencing a decrease of 0.15% percent in the first year following the introduction of the policy, and 0.10% in the second year. In total, from 2014/2015 – 2016/2017, Oestgeest managed a decrease of -0.25%.
On the other hand, Figure 1 shows that Wassenaar was the only municipality that recorded only increases in dropout rates over the same timeframe. In the first year there was an increase of +0.29%, and in the second year, +0.08%. This totals to a +0.37% increase in dropouts over the period of 2014/2015 – 2016/2017.
2014/2015 2015/2016 2016/2017 Kaag & Braassem 1,35% 1,23% 1,74%
Leiden 2,92% 2,32% 2,68% Leiderdorp 1,96% 1,60% 2,16% Oestgeest 1,13% 0,98% 0,88% Voorschoten 1,51% 1,51% 1,17% Wassenaar 1,10% 1,39% 1,47% Zoeterwoude 0,98% 1,30% 0,72% 0,00% 0,50% 1,00% 1,50% 2,00% 2,50% 3,00% 3,50%
Dropout Percentages in SWV VO 2801
Figure 1. Dropout Percentages for the Municipalities within SWV VO 2801. Amended from Onderwijs in Cijfers: VSV in de Gemeenten.
Page | 25 As shown in Figure 2, four other large municipalities within the Netherlands had similar sized
populations of students (≈7000) in comparison to Leiden. These were Alkmaar, Helmond, Oss, and Venlo. All of them, including Leiden, recorded an overall decrease in dropouts from 2014/2015 – 2016/2017. Helmond, Oss, and Venlo followed the same trend as Leiden, experiencing decreases in dropouts the first year after the policy was introduced and minor increases in dropouts the second year. Alkmaar was the only municipality of the group that maintained a decrease in dropouts each year. Leiden’s dropout percentages in Figure 2 are generally higher than the other municipalities, despite having similar population sizes. This could be caused by any number of unknown extraneous variables that fall outside the scope of this thesis.
Based on these dropout percentages and figures, one can determine that in general, SWV VO 2801 has been successful in decreasing its dropout rates since the onset of the policy for suitable education in 2014. That being said, it is difficult to ascertain whether this decrease in dropouts is indeed a
repercussion of the policy networks being formed or not.
According to theory, networks are formed to facilitate the achievement of a common goal (Provan, Fish, & Sydow, 2007). They are goal-oriented (Provan & Lemaire, 2012). If network goals are not achieved in any way, the network could eventually exist in name only (Provan & Kenis, 2007). So, when network goals are achieved, or there is at least some progress in achieving them, networks may be considered effective, especially on the network level. This is the case here, for SWV VO 2801.
2014/2015 2015/2016 2016/2017 Alkmaar 2,25% 1,94% 1,70% Helmond 2,13% 1,91% 2,12% Leiden 2,92% 2,32% 2,68% Oss 1,89% 1,44% 1,60% Venlo 2,20% 1,89% 2,04% 0,00% 0,50% 1,00% 1,50% 2,00% 2,50% 3,00%
3,50%
Dropout Percentages of Similar Major
Municipalities
Figure 2. Dropout Percentages of Major Municipalities with a Comparable Secondary Student Population to Leiden. Amended from Onderwijs in Cijfers: VSV in de Gemeenten.
Page | 26
Subjective Network Effectiveness
Subjective network effectiveness uses the perceptions of network stakeholders to measure how effective the network is. In this case, the researcher determines whether SWV VO 2801 is subjectively effective on the network level by using stakeholder satisfaction, member commitment, and the strength of relationships between and among network members as indicators. The source of information for these indicators were the interviews conducted with some of the network stakeholders, namely the network coordinator and some network board members. The interviews were analysed using a priori coding. Initially, when analysing the interview transcripts, the researcher was looking for data that was relevant to the first level codes, which were determined using theory on network effectiveness. After going through these initial codes again, the researcher was able to go more in depth into the different factors that contributed to these broader themes of subjective network effectiveness. The codes and their frequency of appearance in the data are presented in Table 3. Please be reminded that the full codebooks, which include code descriptions and quote examples directly from the interview reports, may be found in Appendix C.
Table 3. Subjective Network Effectiveness Frequency of Codes
Code (Level 1) Code (Level 2) Frequency
Stakeholder Satisfaction Involvement in Decision-Making 4 Goal Definition 6 Goal Accomplishment 3 Resource Availability 6 Effectiveness 7 Commitment Goal Efficacy 4 High Commitment 11 Low Commitment 2 Strength of Relationships Multiplexity 6 Interorganizational Trust 10 Structural Equivalence 1
Board Member Rotation 1
Outcome Criteria
Stakeholder Satisfaction
As seen in the table, interview respondents described involvement in decision-making, having defined goals, successful goal accomplishment, and resource availability as factors that contributed to their overall satisfaction with the network. All of them described the network as effective when asked, which also contributed to their overall satisfaction with the network.
Because network board members are able to help decide on what is being done within the network, they are usually satisfied with these decisions. Sometimes there are conflicts of interest between decisions made at the network level and the ideologies of the individual schoolboards; however, overall, no decision is made without consent from the network members.
