This Master's thesis investigates the impact of the European Emission Trading System (EU ETS) on Carbon Leakage. It uses regression analysis of panel data to find out whether there is a link between the EU allowance price and trade flows. The study furthermore provides extensive recommendations to policymakers regarding potential improvement of the EU ETS and the overall competitiveness of European Industry.

The problem of female under-representation in STEM still persists, with sexism identified as a significant contributor. Sexism can come in many forms, with gender microaggressions being one of the most prevalent forms in the STEM workplace. Therefore, this thesis explores the phenomenon of gender microaggressions within STEM Energy firms in Greece. By employing a qualitative phenomenological analysis, this research investigates how individual, organizational, and environmental factors contribute to the experiences of gender microaggressions in STEM. Semi-structured interviews with 11 female employees and 2 gender-studies academics were employed. Research was guided by three theoretical frameworks: two distinct taxonomies of gender microaggressions (to investigate the experiences of gender microaggressions), and the Ecological Systems Theory (to investigate the contributing factors). The study resulted in the types of gender microaggressions experienced by women in STEM, their coping mechanisms, and the impact on them. New types of gender microaggressions emerged and were compared to existing literature, leading to the proposal of a new framework, the Integration of Gender Microaggressions taxonomies. This framework provides a new classification of gender microaggressions, integrating the newly emerged types and the theoretical types. Another result of this research is the introduction of the framework of the Ecological Systems Model of Gender Microaggressions. It depicts the dynamic interplay between individual, organizational, and environmental factors that contribute to gender microaggressions. This framework categorizes the contributors into three layers (individual, organizational, and environmental), taking into account the dimension of time. This study provides valuable insights for technological organizations, policymakers, educators and advocacy groups to advance their knowledge about gender microaggressions. It addresses the manifestations of gender microaggressions, women’s coping mechanisms, impact of microaggressions and the contributing factors. The study highlights the pervasiveness of gender microaggressions in the Greek
STEM context, often appearing in more severe manifestations, such as sexual harassment.
By providing a comprehensive understanding of gender microaggressions and their contributors, this
thesis aims to support the well-being of all employees, enhance organizational efficiency, and improve
societal welfare, ultimately contributing to global efforts towards gender equality in the STEM professional
field.

Critical infrastructure systems are vital to societal functions, yet they face increasing complexity due to global challenges such as climate change, urbanization, and digitalization. In the Netherlands, infrastructure administrations play a pivotal role in ensuring the reliability and sustainability of these systems. Interorganizational collaboration (IOC) is recognized as an approach to addressing interdependencies among stakeholders and enhancing infrastructure management. However, there is limited understanding of the organizational preconditions required to foster effective IOC, particularly at the intraorganizational level. This thesis explores the factors that enhance or hinder collaborative capacity among Dutch infrastructure administrations, focusing on the major Port Authority and Railway Administration as case organizations. Using a qualitative, three-step methodology—comprising a literature review, semi-structured interviews, and thematic analysis—it identifies organizational preconditions critical for enabling IOC. Findings highlight that leadership commitment, collaborative goal-setting, structural flexibility, and proactive relationship-building are essential enablers of IOC. Successful collaborations often begin with motivated individuals and are sustained through institutionalized learning and adaptive organizational strategies. The study encourages organizations to balance internal mandates with shared collaborative objectives and to prioritize informal networks and trust-building mechanisms. The findings offer a reflective framework for infrastructure administrations to assess and enhance their collaborative capacity within their own specific context, enabling more effective partnerships in safeguarding critical infrastructure systems. This research contributes to both theory and practice, addressing a gap in understanding how intraorganizational factors shape the readiness and success of interorganizational collaboration.

This thesis investigates the complex dynamics of determinants on span of control within high-tech organizations, with a particular focus on what factors unique to these settings influence span of control. The main research question addressed is: ”How can high-tech organizations determine their span of control?” Through a comprehensive exploratory approach, including qualitative interviews with managers and quantitative data analysis, this research aims to clarify the internal and external factors impacting span of control in high-tech environments. Specifically, it explores how task complexity, team interdependence, geographical factors, and industry specific requirements shape the span of control in ways that differ from traditional industries. The study finds that in high-tech organizations, where task complexity and interdependencies are notably high, a narrower span of control is often required to maintain oversight and quality. However, regional cultural differences also play a critical role, as broader spans are more feasible in high-powerdistance regions due to the cultural acceptance of hierarchical structures. In contrast, low-powerdistance regions tend to favor narrower spans, aligning with preferences for direct managerial oversight. This study also underscores the influence of regulatory requirements, particularly in sectors like semiconductor manufacturing and medical devices, where compliance and precision necessitate close supervision. These factors collectively indicate that span of control in high-tech organizations must be highly adaptable, tailored not only to task demands but also to regional and regulatory contexts. Additionally, the research highlights the potential for technology to enable broader spans of control, albeit with limitations. While automation and digital tools can reduce the need for close supervision in standardized tasks, they do not entirely eliminate the necessity for managerial oversight in complex, interdependent tasks. High-tech managers, particularly those in field service roles, face unique challenges balancing direct customer engagement with team management. This dual responsibility can restrict their capacity to manage broader spans effectively, suggesting a potential role for specialized support roles or digital solutions in handling routine (customer) interactions to allow managers to focus on strategic team oversight. Financial considerations are also prominent, with broader spans offering potential cost savings by reducing the number of managerial layers. However, these benefits must be weighed against operational demands, as broader spans may compromise quality in highly complex environments. Examples of calculations are presented in this study to illustrate possible financial effects of span adjustments. In conclusion, this thesis emphasizes the necessity of a context-sensitive approach to span of control in high-tech environments, where a balance between technological support, regulatory requirements, and cultural influences is crucial. Recommendations for future research include longitudinal studies to observe how technological advancements might further impact span of control over time and expanding sample diversity to encompass a broader range of high-tech sectors. This research contributes to the field by providing a nuanced framework for understanding span of control in complex, dynamic settings, offering practical insights for high-tech organizations.

The airline industry is going through a big transformation. The institutional environment as well as the industry itself has changed. Competition with the entry of the low-cost carriers in a market that used to be dominated by flagship carriers and societal pressure regarding sustainability on airlines with COVID-19 state aids increased, while the shortage in the labour market limited the pace of recovery. The technologies that airlines use to operate have been outdated to cope with today's massive challenges. All of these changes forced airlines to innovate to maintain their services. During the era of technology, these innovations can be better nurtured within an innovation-conducive environment, which might be seen as an ecosystem designed for the co-creation of value through collaboration. In the last decade, both managers and scholars showed an increasing interest in the concept of innovation ecosystems, which provides a holistic perspective for value creation. However, despite its potential, the implementation of the innovation ecosystems in the airline industry and how it can support an airline to implement its innovation strategy has not been investigated. By conducting literature research and qualitative embedded case study, this research concludes that implementation of innovation ecosystems and innovation ecosystem strategies can help an airline to achieve its objectives on innovation and implement its innovation strategy by aligning its innovation processes, overcoming the barriers to innovation that hampers its innovation processes and finally facilitating the collaboration within and beyond the value chain to develop radical innovations that are required to cope with today’s complex challenges.

The “Wet open overheid” or Woo is a law that facilitates access to government information for Dutch citizens and states that “everyone has the right to access government information without having to show an interest in it, subject to the limitations provided for by this law.” This law is the Dutch embodiment of a Freedom of Information (FOI) Act. Facilitating citizens with access to government information is an essential promotor of an open and democratic society. Furthermore, it allows citizens to hold their governments accountable for their actions and make informed political decisions. The proper execution of this right to government information is essential, however, it was found that processing the information requests directed to Ministries of the Dutch Government takes an average of 161 days, where the law states a maximum handling term of 42 days. This delay impedes the citizens’ access to government information and should be addressed. While previous research has been performed towards improving “Woo-request processing,” there is little known about the process itself, nor what the direct causes of delay are. Information management is suggested as a possible culprit. Yet, there is little known about its effects, which are likely to be noticeable in the information search process (ISP) that is part of Woo-request processing. Therefore, the objective of this research is to identify how the existing Woo-request ISP is executed at Ministries of the Dutch National Government and what influences it in order to define improvements for said process in relation to its facilitation of an open and transparent government.

This research concludes that the Woo-request ISP is a complex process that is influenced by many factors. Whereas the ISP described by academic literature portrays multiple iterative steps, the Woo-request ISP hardly portrays any iterative steps in relation to an improved definition and understanding of the information problem. As a result, the requestor cannot learn from the search itself and the information found throughout it. Combined with a possible lack of prerequisite knowledge of the requestor, this means that requests are likely to remain broadly defined and vague due to a fear of missing out on information caused by the requestor’s distrust. Therefore, this research suggests that through increased informative interaction between the requestor and the handler, supported by search and content experts, specification of the requests should be facilitated, which is expected to improve the search process. This also builds a relationship of trust between the parties involved, which is expected to increase the chance of a requestor’s cooperation. Furthermore, the Woo-request ISP as it is does not guarantee completeness nor traceability of the search performed. Therefore it is suggested that the information locating phase makes use of a metadata-based centralized search tool that is operated by experienced and trained personnel who are able to access all information applicable to the Woo.

Whilst the suggested improvements are likely to have a positive influence on the existing process; there are large hurdles to be overcome. It must be noted that some problems or dependencies might in fact never be overcome. For example, the realization of a centrally accessible archive might never see the day of light, as it conflicts with government tasks and values other than openness and transparency. Additionally, it must be noted that the existing “document definition” used by the Dutch government is incredibly broad. This broad definition encases that if the government is to adhere to its laws, almost all currently produced information must be searchable by Woo-request handlers. Especially in a government where a policy rarely stands on its own, it can be challenging to define the end of relevance to a request. Lastly, it must be noted that some requestors might never be pleased with the outcomes of their Woo-request.

This Master Thesis project was the last assignment in obtaining a MSc degree in Science Communication at the Delft University of Technology. The focus of the project was on finding fitting leadership styles, structures, or behaviors that stimulate learning and help develop a learning mindset in a transdisciplinary collaboration context.To deal with the increasing complexity and urgency of societal issues – such as the aging population, climate change, and the energy transition – a new collaboration revolution has taken shape. This revolution, also referred to as convergence, is still very new. Articles available on what convergence, and the transdisciplinary collaboration as the new collaboration type therein entail, are rather vague and abstract. The HOW of the matter is not yet clear. Learning is inextricably linked to transdisciplinary collaboration due to its novelty. However, explicit attention has not been given to building in learning and a learning mindset in relation to transdisciplinary collaboration. Leadership and learning have had an existing link in literature, but have not been regarded in a setting of transdisciplinary collaboration. Taking leadership into account will enable the formulation and implementation of more concrete actions to move away from a still rather vague idea of transdisciplinary collaboration and towards actionable steps. Therefore, the combination of transdisciplinary collaboration, learning, and leadership is not only interesting and relevant from a research point of view but it will also directly influence the effectivity of convergence endeavors of the Erasmus Medical Centre (EMC), Erasmus University of Rotterdam (EUR), and Delft University of Technology (TUD) that are currently taking place. Research on the relation between transdisciplinary collaboration, learning, and leadership has not been conducted, and thus forms the main focus of this Master Thesis project. The main research question therefore is: What leadership styles, structures, or behaviors, suited for the transdisciplinary collaboration context of the Digital Twin project of the Health & Technology branch of the Convergence Agenda of the Erasmus Medical Centre, the Erasmus University Rotterdam, and the Delft University of Technology, stimulate and facilitate learning and help develop a learning mindset?Based on an extensive literature study, a qualitative study composed of semi-structured interviews with people involved in My Digital Twin – one of the flagships of the Health & Technology theme of the Convergence agenda of the TUD, EUR, and EMC – and analysis of the qualitative data using structural coding in Atlas Ti 9, core elements of transdisciplinary collaboration and core elements of learning in a transdisciplinary collaboration context were found. Based on these elements, a long list of appropriate behaviors for leaders to portray to stimulate learning or work towards transdisciplinarity was established. Appropriate leadership styles and structures were concluded to be rotational distributed leadership consisting of formal and informal leaders. The discussion shows ideas for tools to stimulate learning and help develop a learning mindset, discusses the quality of the research, and provides recommendations for future research.Key words: transdisciplinarity, transdisciplinary collaboration, learning, leadership, duality, distributed leadership, convergence

In 2011, Adaptive Delta Management (ADM) was introduced in the Dutch Delta Program as a policy development method to incorporate uncertainty in decision making transparently. Currently, little is known over the functioning of ADM in practice, while this is determining its success. The study explores the application of ADM in Dutch practice. Starting from the scientific foundation of ADM is being researched how ADM is shaped in practice using interviews with actors on the strategy level and the implementation level. The results are analyzed to explore what they mean for the adaptiveness of coping with sea-level rise by Dutch delta management. The research shows that the application of the greater part of the instruments of Adaptive Delta Management (ADM) is not coherent with how they should be applied according to theory. The first difference observed is that adaptation pathway maps in practice are far simpler than prescribed in theory; they only contain the preferential strategy instead of multiple strategies, and no signposts and transfer stations are defined. Secondly, adaptation tipping points are more flexible in practice than is prescribed in theory. The final difference observed is that evaluating strategies in practice has a fixed rhythm, while according to theory, continuous evaluation and recalibration should occur whenever new information comes available. The main implication of these differences is that ADM in practice provides less guidance to policymakers on when and which adjustments of strategies are needed than in theory. Based on the analysis of the similarities and differences between theory and practice, the research found that currently, the governance of ADM in practice in the Netherlands does not enhance adaptation to sea-level rise. The study stresses the importance of taking action to keep the Netherlands safe from sea-level rise now and in the future. Therefore, several recommendations inspired by the theoretical foundation of ADM are proposed to ensure adaptation to sea-level rise is enhanced. The first recommendation aims to ensure that policymakers get insight into the solution space to cope with sea-level rise and understand what measures need to be taken today to keep the long-term solutions open. Furthermore, actors involved in the implementation of strategies (such as regional water authorities, municipalities and water utility companies) should obtain insight into the possible consequences of sea-level rise for their area and know how to incorporate this in decision making.

In 2019, the Dutch nitrogen crisis emerged after the Dutch legislation on nitrogen pollution (PAS) was dismissed by the Dutch Raad van State. The crisis resulted in large scale demonstration by farmers and lingering uncertainty in the Dutch business sector. From an ecological perspective the current policy approach is limited as it focusses primarily on reducing the inflow of nitrogen to Natura 2000 areas. Widening the policy approach to better represent the ecological side of the nitrogen crisis provides alternative ecological solution avenues to resolve the crisis. To this end, a System Dynamics model is constructed that encompasses the nutrient cycles of N and P, as well as a level of ecological complexity. To investigate the model’s behaviour under circumstances of deep uncertainty an Exploratory Modelling and Analysis (EMA) approach is used. The results indicate a large potential for cost reduction and increase of ecological performance by aiming policy at ecological compensation instead of emission reduction. The implementation of ecological measures requires scientific substantiation to be accepted into the regulatory framework. The potential economic and societal benefits of an ecological policy approach should be a strong incentive for policy makers and scientists to develop the empirical knowledge for ecological engineering.

Situation: A large portion of the beneficiaries of humanitarian aid have little to no proof of their identity, this is especially the case in African countries, in which humanitarian intervention is common. In order to facilitate aid to these individuals, humanitarian organizations (HOs) leverage their on-the-ground capacity to create identity profiles and risk assessments of these people. However, in-kind aid is being increasingly replaced by Cash Transfer Programs (CTPs). In CTPs, beneficiaries are provided with funding to self-procure their necessities. Complication: For efficient CTPs, these humanitarian identities need to be accepted beyond the boundaries of humanitarian aid. However, by relying on traditional identity management systems, HOs expose the beneficiaries to security and function creep risks. In order to share beneficiary identity information in a more responsible way, HOs have started to leverage Self-Sovereign Identity (SSI) systems. These SSI systems need to be scaled up to a more foundational nature in order to facilitate private-sector services. For this a collaboration with national public- and private stakeholders is required. In order to realize that, there is a need for a process design. Which among other things ensures support from crucial stakeholders. Approach: This study used a Design Science Research inspired approach, combined with a Systems Engineering perspective to explicate the problem, define support driving and constraining circumstances and conditions, generate principles for the humanitarian sector to create support driving circumstances and conditions in Kenya and validate these principles using the input of industry experts. This research approach had a focus of practical insights over theoretical insights. Results: A set of five validated support nurturing principles were established with which humanitarian organizations can nurture support for a more foundational humanitarian SSI system in Kenya. Additionally, a framework has been composed with which local circumstances and conditions in a country can be assessed. Next steps: Further research should focus on establishing a more complete process design for a collaboration process, which also deals with participation of stakeholders, structures commitment and defines process rules for different phases of the process. Additionally, further research should explore the capacity and willingness of beneficiaries to control their own identity. And finally, the effect of international innovation initiation on the willingness of national stakeholders should be explored.

Disinformation is a problem that has become more relevant over the past years. Also the European Union tried and is trying to deal with this problem by introducing multiple policy actions. However, the impact and usefulness of these policies is not clear. This thesis uses a combination of stakeholder analysis, interviews and systems thinking to assess the impact of current policies the reduce the spread and production of disinformation. This assessment serves to further improve current and future policy actions. Results from the analysis indicate that current policy actions are not focused on the root causes of the production and spread, and can be improved, among other things, by introducing regulation, making policies more goal-oriented and being more clear about policies in terms of content.

Introduction. To contribute to the reduction of global warming, The Netherlands signed the Paris Agreement on 12-12-2015, which dictates that in 2030, a reduction of 55% in CO¬2 emissions compared to 1990 must be achieved (P. Boot et al., 2016; Sociaal-Economische Raad, 2013). To meet these goals, several Dutch regions started their Regional Energy Strategy (RES) formulation process in 2016. A RES is a collaboration of private and public parties that strives to effectively balance energy supply and demand within the regional boundaries by focussing on sustainable energy generation, energy savings and the benefits from economies of scale (Schuurs & Schwencke, 2017). Going through such a process with the relevant stakeholders will ultimately result in new regional policy in the field of energy. The ‘Metropolitan region Rotterdam The Hague’ (MRDH) is one of the 30 regions of the Netherlands. Its diverse economic activity, great variety in municipality sizes and key economic and infrastructural players such as the Port of Rotterdam and Greenport West-Holland contribute to the region’s complexity. The MRDH started the RES formulation process in February 2018 and endeavours to deliver the outcome in July 2019. The owners of this process are its 23 municipalities, its waterboards and the province of South-Holland. 
Knowledge gap. In the field of process management, several core values exist which play a role in each process, such as trust, openness, progress, and content. Scientific literature does not express the weight of these values, as these depend on the process’ context. Additionally, it is unclear how a high level of complexity impacts the process architecture and potential process interventions. In the field of regional governance very limited scientific literature is available. That in itself is remarkable, as the process approach is growing steadily (Boogers, Klok, Denters, & Sanders, 2016). Regional governance –a governance level in between the municipal and provincial level– is controversial. The scientific debate focuses on several aspects. First, regional governance would lead to administrative chaos and complexity. Second, regional governance could lead to the hollowing out of local governance, thereby reducing the democratic quality. Lastly, the question is what organisational form of regional governance can best be employed; is it monocentric, a concentration of regional governance in one authority, or polycentric, a fragmented system of collaborative arrangements into which independent municipalities are free to enter?
Central research question. The process’ stakeholders have different perceptions of the RES assignment. These ‘different assignment perceptions’ are caused by the abstractness of the assignment, and the fact that the original assignment has been changed by the influences of the interim reports of the Climate Agreement. The Climate Agreement is a national treaty that mainly aims at the reduction of CO2 emissions in the Netherlands. The fact that the stakeholders perceive the assignment differently leads to several problems in the formulation process. Some of these problems even form a severe threat to the core values of process management. To solve these problems, the process managers perform a variety of process interventions to align the perceptions of the stakeholders.The main research question is: ‘How are problems resulting from stakeholders’ different perceptions of the assignment ‘Energiestrategie regio Rotterdam Den Haag’ addressed by the use of process management?Research design and methods. This study applies the case study method on the MRDH region. To answer the research questions, both qualitative and quantitative methods are used. First, an examination is made of what the leading assignment for the RES MRDH is and how it has been influenced by the Climate Agreement. Second, the assignment perceptions of the owners are examined through 26 stakeholder interviews. The parties that published the RES assignment (i.e., IPO, VNG, UvW and the Ministry of Internal Affairs) are interviewed on their perception as well. These data are analysed using the qualitative data software ‘ATLAS TI’. Through a workshop with the process managers it is determined how the different assignment perceptions impact on the process. Utilising an influence tree, the consequences of these different assignment perceptions are visualised. It has turned out that real problems that actually occurred in the process could be linked to the consequences. Next, the process coordinator has been interviewed several times to analyse the context and the performed process interventions of these problems through a lens of process management. The last part describes the currently relevant problems that still need process intervention(s). For one problem, a ‘solution-panel’ has been gathered, consisting of several experts, to find suitable process interventions. 
Results. The findings show that the interim deliverables of the Climate Agreement demand a higher level of detail of the RES outcome. The interviews indicate that the majority of the owners share the following image: ‘The RES is a municipal and regional plan that is based on facts and figures, intended to work towards a future energy system in pursuit of the regional or national climate objectives’ (Section 5.5.1, Figure 7). The assignment perception of the Ministry of Internal Affairs and the umbrella organisations comes down to: ‘The RES is a masterplan which combines multiple perspectives. Eventually, these different dimensions must be merged and realised in the municipalities’ spatial environment’ (Section 5.7). Concerning process management, the process architect and coordinator have chosen to implement two foundations for a robust process course. These provide solidity, controllability and should prevent problems. Nevertheless, ten problems have been identified as a result of different assignment perceptions. Six of them have been resolved by conducting municipal consultations and the extension of the process lead time. These process interventions have been performed to secure an adequate involvement of the municipalities’ official organisation to the RES formulation process. Another process intervention included the reframing of the RES to meet the expectations of specific stakeholders. Next, to stimulate the spatial integration of the assignment, the spatial area civil servants were invited for the workshops during the spring of 2019. Lastly, bilateral meetings were held between the process coordinator and the aldermen of the municipalities to clarify points of view and to adjust the aldermen their attitude in the RES process. To solve the problem of ‘the national RES assignment demands for concrete top-down objectives while the RES needs bottom-up collaboration’, the solution panel has suggested that process management can be employed for creating a safe environment in the pursuit of ‘shared ownership’ of the assignment.
Discussion. An interpretation of the applied process management is that the high complexity of the region has led to a wide range of assignment perceptions which in turn needed comprehensive process architecture and several process interventions. Besides, the process managers value the core values of process management of De Bruijn, Ten Heuvelhof, & In ’T Veld (2010) depending upon the process agreements and the agreed level of ambition. Also, the results led to the assumption that the core value ‘progress’ forms a trade-off with ‘content’. Concerning regional governance; ‘administrative chaos and complexity’ carries a negative emotional charge. The popular thought is that administrative chaos and complexity causes reduced democratic legitimacy, less trust due to the multitude of partners, and poor effectiveness. These are incorrect by fact. Therefore, the researcher opts for new terminology: ‘administrative intensification’. The results of the democratic quality of the RES can be interpreted as excellent, as only two of the 23 municipalities have no (or partial) influence on the RES formulation process. In fact, these two municipalities were not represented as they had internal problems. Based on this result, the findings of Boogers et al. (2016) –indicating that around 90% of the alderman having an effective influence on regional decision-making– can be confirmed. Further research. Regarding process management, further research should examine the RES formulation processes of the other Dutch regions to discover how the core values of process management have been applied, and whether relations between the process context and the core values exist. In the field of regional governance, further research should investigate how this ‘new renewable energy implementation challenge’ can be poured into the current multi-layer governance system while preserving democratic legitimacy. Furthermore, when regional governance structures resulting from the RES process have been set up, further research should be undertaken to investigate the effectiveness of the collaborations in such structure.

Simulation models are increasingly used to substantiate long-term planning policy decisions. The communication of model results from modellers to policy-makers has not received much attention by scholars. While, without correct interpretation and correct use of model results the quality of the model itself is irrelevant for the impact of the model on policy. This research analyses this gap in literature by looking at a system from model to policy, using the case of the 2017 office policy of Amsterdam.

Agility is increasingly being used in the public sector as its adaptive approach through short-cycled development could allow public services to keep up with developments in technology and the complex problems contemporary societies face. There are, however, some dilemmas at play when it comes to agility and the public sector’s transparency and stability for accountability. Furthermore, agile projects appear to be less predictable and more likely to go off track. The main paradox of agility is that even though it appears to be at odds with control, it is required for agility to be effective. Besides, public sector bureaucracies come with their own control mechanisms which theoretically conflict with agility. The research explores these phenomena by considering how control over agile processes in the public sector can be enhanced through the use of control mechanisms. By using an explorative and single case study approach with validation sessions the research combines qualitative data from theoretical and practical findings which are analyzed through a Revised Levers of Control framework to provide recommendations for controlling public sector agility. It indicates that the agile process itself is the main means of control for which an extensive list of mechanisms is provided of both strategic and operational nature. Besides, the vast presence of interfaces in the public sector increases complexity while external deadlines and tight budgets provide a clear desire for planning as control. A main recommendation therefore consists of mechanisms to be used to perform planning within agility. The research also showcases that types of bureaucratic control can facilitate or are even required for the agile process to function when it comes to centralized strategic visions on priorities and an organizational vision on implementation of agility. Overall, this implies control over agility in the public sector to be a balancing act between mechanisms associated with ‘pure agility’ and with public sector bureaucracies. Further empirical research into the use and implications of proposed control mechanisms and presence of interfaces in practice would help in enriching the findings and provide more insight into the requirements of different types of public sector organizations to control their agile processes.