Standard setting the Dutch way

England is moving away from work being marked by teachers in schools. But in the Netherlands, teachers remain a trusted part of the assessment system, with no talk of grade inflation. Anton Béguin explains how it works.

Hand writing grade on paperHalf of the central examinations for secondary education in the Netherlands are completely school-based – they are both constructed and marked by the school. The other half are examinations that are constructed centrally, but still marked in schools, with second marking done by a teacher in another school. If there is disagreement, the two teachers discuss the results and if they cannot come to an agreement they take the average score.

The system in England for the reformed GCSEs couldn’t be further from this model. The reform of GCSEs is being delivered from the centre outwards. At the centre are experts appointed by the Secretary of State for Education, who set out priorities for education across the country. These priorities are translated into programmes of study by the Department for Education. Ofqual, the examinations regulator, then organises the detail of what skills and knowledge should be assessed in the examinations. Finally, the awarding bodies draw up syllabuses and examinations that will meet the government’s criteria.

At every stage of the design process there is intense public scrutiny and rigorous evaluation, but there is little direct involvement on the part of schools. Once the reformed examinations are up and running, it is extremely unlikely that school teachers will be marking any of the work of their own students.

Imagine instead giving schools direct responsibility to design and mark half of the assessments, and the responsibility to mark the other half based on a central marking scheme. This seems unlikely to be an idea that the examinations regulator would welcome, given the recent court battle over standards in GCSE English.

So how do we do things so differently in the Netherlands, and without public furore? Probably the most important ingredient is the greater involvement of schools. In the examination system there is more balance between centrally-based and school-based functions. There are statistical controls in place to inform the standard-setting, but the schools have far more responsibility and are far more involved than they are in England.

An aspect common to both systems is that before 2000 in the Netherlands, the central board only adjusted marks when it suspected an examination was harder, not when it was easier. There is some evidence that examiners in England in the past also tended to adjust grade boundaries only when then they suspected an examination was too hard. Ofqual’s recent experience suggests that defending an increase in grade boundaries is far from easy.

So has there really been no inflation in the scores in the Dutch system over time? I have not noticed any yet. Of course, if you were able to make a reliable, detailed comparison of standards over time, you would find some respects in which Dutch students used to perform better than they do now, and others where they used to perform worse. Whether they perform better overall is probably more a question of what is valued in the curriculum than a reflection of any absolute standard.

Anton Béguin is director of research at Cito, a testing and assessment company based in the Netherlands, and a member of the CERP Advisory Group.

Share this page