Since the turn of the century, Portugal has been experiencing the emergence of school administration models emphasizing community participation, accompanied by a shift towards granting greater autonomy to schools. This shift is aligned with the principles of the New Public Management's ‘educational toolkit’ (Verger & Curran, 2014, p.256). As a result of this transformation, diverse programs and practices have been introduced, with a strong emphasis on school-based management and pedagogy. Schools are now empowered to make decisions regarding their curriculum and educational initiatives, tailored to their specific social contexts (ibid). This is demonstrated through a series of policies implemented through reflection, negotiation, and collaboration, such as school external evaluation. These policies have resulted in a gradual reduction of hierarchical control by the State and have paved the way for network governance. One example is the 'Pedagogical Innovation Pilot Project' (PIPP) (2016-2019), an initiative based on school-based management that aims to promote student success and address school dropout rates.

PIPP was implemented in six school clusters nationwide, providing participating schools with increased autonomy in organizational, pedagogical, and curricular areas (Costa & Almeida, 2019). It involved approximately 744 teachers and 7,844 students across various grades and locations to enhance the quality of learning and educational outcomes. Additionally, it aimed to tackle the issues of school dropout and failure across all teaching cycles by facilitating the implementation of innovation projects in participating schools (Portugal, 2017).

The outcomes of PPIP of reducing school dropout and improving retention rates have been very positive. These achievements have been realized by modifying the rigid pedagogical structure of schools and reshaping the perceptions of school actors regarding how school, curriculum, and student assessment should be delivered. Additionally, PPIP has instigated a recent policy that grants schools the authority to manage more than 25% of the national curriculum, subject to an innovation plan developed by the schools and approved by the Ministry of Education (ME) (Portugal, 2019).

This paper builds upon a previous study (Carvalho, Costa & Almeida, 2020) that emphasized the importance of knowledge in policy-politics and the underlying logic of PPIP, where the production, legitimization, and dissemination of knowledge facilitate the coordination and control of actions in the educational field. We aim to examine PIPP as a lens to analyze the shifts in the steering of the education system, specifically through the perspective of network governance. This leads to the research question: How does PIPP exemplify the rise of network-based coordination and control within the education system? To address this question, three objectives were established: (i) to identify and analyze the actors involved in the design and implementation of PIPP, (ii) to examine the instruments and forms of control utilized, and (iii) to analyze the interactions among the involved actors.

The study is grounded in the public action approach to public policies (Lascoumes & Le Galès, 2007; Hassenteufel, 2008), which emphasizes the involvement of various actors in shaping and interpreting activities within the public sphere of education, extending beyond government intervention (Van Zanten, 2000). To achieve this, we will employ the concept of governance, which considers the relationship between State intervention and societal autonomy, spanning a continuum from public authority to societal self-regulation (Treib, Bähr & Falkner, 2007; Barroso, 2005). Drawing upon Meuleman's (2008) typology of hierarchical, network, and market governance, which can manifest in different combinations, our objective is to comprehend the factors that hinder the shift toward network steering and examine the implications of novel governance approaches, particularly about the coexistence with hierarchical governance.

A qualitative research methodology was used based on an interpretive approach (Cohen, Manion & Morrison, 2007). Archival research techniques and interviews with key actors (such as school and deputy principals, coordinators, and class directors) were conducted, guided by the assumption that these documents represent tangible outcomes facilitating cooperation among the various actors (Carvalho, 2006, p.42). The analysis of documents encompassed a wide range of materials, including legislation and official internal documents from the ME and the government, monitoring reports, school projects, evaluation reports, as well as agendas of meetings and seminars. The interviews were conducted with 86 key informants and comprised two types: semi-structured and focus group interviews. The semi-structured interviews involved one high official from the ME, six school principals, one deputy principal, and three assistant principals. Additionally, 13 interviews were conducted with PIPP coordinators at the school clusters. Focus group interviews were held with 37 middle managers, including department coordinators, general coordinators, and psychologists, as well as 25-year/class coordinators. The data from the interviews were transcribed and analyzed using the methodology outlined by Bardin (2009). Both the written documents and interviews enabled us to address the research objectives by identifying and analyzing the actors involved in the design and implementation of the PIPP. Additionally, we examined the instruments and types of control used by these actors and explored how they interact with each other. For the analysis of documents and interviews, we utilized the deductive method, employing the categories of analysis "who" (actors) (state/non-state) and "how" (meetings /seminars) of the PIPP. Additionally, we drew inspiration from two specific features of governance based on Meuleman's (2008) dimensions to guide the categories and subcategories of analysis. These features include the types of instruments utilized, such as legislation/compliance (associated with hierarchical governance) or voluntary/contracting instruments that require the actors' adherence (associated with network governance). We also considered the way control was established, namely, through authority processes (hierarchical, top-down) or based on trust (horizontal, networked, resulting from goal consensus).

PIPP represents a shift in the modes of coordination employed by state authorities, aiming to eliminate or minimize retention and dropout rates through network governance and the engagement of diverse actors. While this intent was successfully achieved, there still existed a dominance of the 'Rule of Law' and Control, a fundamental aspect of public administration accomplished by hierarchy (Hood, 1991). State authorities willingly relinquished some formal authority, entrusting schools to make their own decisions and assume responsibilities. While the relationship between state authorities and schools followed a vertical structure in terms of project design, requiring validation and ongoing monitoring by the Ministry, emphasis was placed on nurturing horizontal relationships. The Ministry of Education (ME) actively engaged schools in meditative and reflective activities, which principals then implemented in their schools. Networking played a crucial role, with events highly valued for facilitating collaborative interaction, a key feature of governance networks (Tenbensel, 2005). The ME played a significant role in managing the network, organizing meetings and seminars, and demonstrating concern for participants' needs. This contributed to the development of routine interaction, which is critical for maintaining and building trust among participants (McEvily & Zaheer, 2004). Trust, unlike hierarchy, is not based on formal control but on dependency and earlier interactions, core features of governance networks (Rousseau, Sitkin, Burt & Camerer, 1998; Klijn, 2010). The coexistence of new governance modes alongside hierarchy presents challenges (Héritier, 2003; Eberlein & Kerwer, 2004), and public policy processes often require different governance styles to accommodate diverse phases (Meuleman, 2011). In the case of PIPP, schools expected the state to remain closely involved and supportive, with the state playing a key brokering role as an intermediary between the national and local levels.

Bardin, L. (2009). Análise de conteúdo. Lisboa: Edições 70. Barroso, J. (2005). O Estado, a educação e a regulação das políticas públicas. E&S, 26 (92), pp. 725-751. Carvalho, L. M. (2006). Apontamentos sobre as relações entre conhecimento e política educativa. Administração Educacional (6), pp. 36-45. Carvalho, L.M., Costa, E., & Almeida, M. (2021). Recontextualization of improvement-oriented policies in Portugal: the case of the PPIP (2016-2019). IJER. https://doi.org/10.1016/j.ijer.2021.101865 Cohen, L., Manion, L., & Morrison, K. (2007) Research methods in education. London: Routledge. Costa, E., & Almeida, M. (2019). Evaluation Study of the Pedagogical Innovation Pilot Project. Lisboa: IE-ULisboa/MEC/DGE. Eberlein, B., Kerwer, D. (2004) ‘New Governance in the European Union: A Theoretical Perspective’, JCMS 42(1): 121–42. Hassenteufel, P. (2008). Sociologie politique: l’action publique. Paris: AC. Héritier, A. (2003) New Modes of Governance in Europe: Increasing Political Capacity and Policy Effectiveness. The State of the European Union, 6. Oxford: UP. Klijn, E. H. (2010). Trust in Governance Networks: Looking for Conditions for Innovative Solutions and Outcomes. In The new public governance?. NY: Routledge. Lascoumes, P., & Le Galès, P. (2007). Sociologie de l’action publique. Paris: AC. McEvily, B., & Zaheer, A. (2004). Architects of trust: The role of network facilitators in geographical clusters. In Trust and Distrust in Organizations (pp. 189-213). RSF. Meuleman, L. (2008). Public Management and the Metagovernance of Hierarchies, Networks and Markets. Dordrecht: Springer. Meuleman, L. (2011). Metagoverning governance styles–broadening the public manager's action. In Interactive Policymaking, Metagovernance and Democracy (pp. 95–110). HQ, UK: ECPR Press. PORTUGAL (2017). Despacho n.º 3721/2017 - DR n.º 85/2017, Série II de 2017-05-03 https://dre.pt/application/conteudo/106958832 PORTUGAL (2019). Portaria nº 181/2019, DR n.º 111/2019, Série I de 2019-06-11, https://dre.pt/home/-/dre/122541299/details/maximized Rousseau D., Sitkin S. B., Burt R. S., Camerer C. (1998). Not so different after all: A cross discipline view of trust. The AMR, 23, 393-404. Tenbensel, T. (2005) Multiple modes of governance, Public Management Review, 7:2, 267-288, https://www.tandfonline.com/doi/abs/10.1080/14719030500091566 Treib, O., Bähr, H., & Falkner, G. (2007) Modes of governance: towards a conceptual clarification, Journal European Public Policy, 14:1, 1-20, https://doi.org/10.1080/135017606061071406 Van Zanten, A. (2000). Les Politiques de l’Éducation. Paris: PUF. Verger, A., & Curran, M. (2014). NPM as a global education policy: its adoption and re-contextualization in a Southern European setting, Critical Studies in Education, 55:3, 253-271, https://doi.org/10.1080/17508487.2014.913531 Windzio, M., Sackmann, R., & Martens, K. (2005). Types of Governance in Education – A Quantitative Analysis. Bremen: Sfb 597 „Staatlichkeit im Wandel“