Join us in Seattle on June 26-29 for the NCEE Leaders Retreat. Learn more here.

By: Jackie Kraemer

OECD report coverThe OECD’s Centre for Educational Research and Innovation (CERI) released the first of a series of planned case studies that are part of its Governing Complex Education Systems (GCES) project.  The purpose of GCES is to explore what models of governance are effective in complex educational systems.  The key areas the project is focusing on are: accountability; capacity building; and policy design and implementation.  GCES has developed analytical papers on this topic; is supporting a series of case studies and has held a series of thematic conferences on governance from the Centre, from the local level and at multi-levels. The initial phase of the project focused on establishing the analytical basis for the work with the next two years focusing more specifically on issues around accountability, capacity building and strategic thinking on a systems level.

In the project’s first case study, Coping with very weak primary schools: Towards smart interventions in Dutch education policy, published in July, OECD explores decentralization of school operations.  The case study offers a useful perspective on how to achieve national objectives in a decentralized education system and the trade-offs involved in focusing resources only on the most needy schools.  The Netherlands was chosen as an extreme case of decentralized governance in education, as Dutch schools are completely autonomous operating only under the framework set out by the national Ministry of Education.  Twelve hundred school boards manage over 7,000 schools in the country with no level of oversight between the school boards and the Ministry.

The case focuses, in particular, on how this system deals with weak primary schools.  Schools in the Netherlands are inspected regularly.  Among its functions, the inspectorate assesses the “risk” of each underperforming school.  The risk assessment takes into consideration outputs such as grades and graduation rates.  Schools that are deemed at high risk then receive a full inspection and receive one of three rankings (normal, weak or very weak).  Schools that are labeled weak or very weak are “supervised” by the inspectorate for the next two years and are “offered” services that are delivered through the Council for Primary Education, which was established in 2007.  The Council was created to separate the oversight and service functions of the inspectorate and especially to mitigate against conflicts of interest.

Low performing schools can receive a variety of services offered by the Council including: a gap analysis to identify weaknesses and a report making specific recommendations on how to fix those weaknesses; expert teams that stay on-site to support the schools as they work on improvement strategies; and “twinning” projects to pair high performing schools with underperforming schools.

Schools that are functioning well based on an inspection are not interfered with.  This allows the Netherlands to spend resources primarily on schools that need help.

The case study concludes that this system works in the Netherlands, as almost all weak or very weak schools improve considerably and feedback from the schools on the types of support they receive are good.  The paper does, however, raise some points about this model of inspection that are worth considering:

  • Even though most of the schools labeled weak and very weak (less than one percent) improve, the system does allow children to remain in the underperforming school for two years.
  • The model of focusing in on schools only when they are weak or very weak does nothing to keep schools from becoming weak.  There is little focus on prevention of failure or to stimulate excellence in schools that are performing “normally”.
  • While the Dutch have now separated the role of inspecting the schools from assisting the schools in need, the study found that the understanding of this by the schools is not completely clear.
  • The study also notes the importance of studying the contexts in which specific interventions work well and not well, as all interventions do not work well in all schools.  Moving from what the report characterizes as a “linear” to a “circular” model of intervention that assesses and reassesses a turnaround strategy at many steps along the way would help to develop “smarter” interventions better adapted to the unique circumstances at any particular school.


DutchprimarystudentsThe Dutch case provides one approach to targeting limited resources in an effective way, with comprehensive interventions focused only on the most needy schools for limited periods of time.  Dutch schools have very little monitoring outside of the inspection process.  However, readers should note that the Ministry of Education, through the inspectorate and with the help of the Council, does hold primary schools accountable for their performance so that the system is not entirely driven by local decision makers.

OECD’s Governing Complex Education Systems project is planning to release two additional case studies this year.  One will report on a new school inspection system in Poland while the other will discuss the implementation of the Assessment for Learning system in Norway which revised how formative assessment is done in Norwegian classrooms.  The Norway case examines how a national reform is implemented in another decentralized system with high levels of teacher autonomy.  It was picked specifically to contrast a national reform with the targeted interventions analyzed in the Netherlands case.

For information about the history of reform in high performing countries, visit