Benchmarks or Competitive Assessment?

Imagine that you’re on an operating table and you somehow learnt that your surgeon is an ‘A’ grade student, but their ‘real’ scores in their final surgeon’s exam was actually 45% (which means the rest of his class did worse than this). Would you be happy to know that you’re about to be operated on by a 45% exam scoring surgeon? Would you assume that the reason for the low marks was that the examination set for the relevant year was too hard? Would you instead consider the possibility that the whole class were poor learners, or perhaps they had a terrible lecturer/demonstrator? I know I would err on the side of caution and ask for another surgeon. What has this to do with education?

Examinations appear to be a rite of passage for the vast majority of learners.  However, the exam (for better of for worse) is an integral part of the education system that determines one’s passage through the educational system and finally one’s career prospects. I still have nightmares about turning up to examinations late, or completely forgetting anything that I thought I learned about a subject, and this was for exams that I took decades ago. Clearly examinations are incredibly stressful and in an extreme case may sadly end in suicide. The topic of how, when why and where examinations are held, is worth taking seriously.

Perhaps what is not so well understood by the everyday public is what happens both before and then after the actual examination period. Before the examination is taken, the people who set the exam try to write one that does it’s best to differentiate between the best and the worst students in the curriculum topic. So under this system, you do not want to write an exam in which almost all the children get the answers ‘right’. You want the best children to get the answer right and the worst to get it wrong.

After the exam when scripts are being marked there is another process that goes on to ‘help’ or ensure that the children being examined can be graded from ‘best’ to ‘worst’. It is called ‘normalising’ the results. It is assumed that the top scorers would in always get an ‘A’ grade’ marks (say 85% or more). If the ‘top scorers’  had scores that were less than this marks, then it is assumed that the examination set, was too hard. Everyone’s scores are then adjusted to reflect the usual range of marks. This is one of the reasons that examinations are never returned because otherwise this ‘adjusting’ of the results would be more obvious.

The consequences of this are threefold: 

  • firstly, the examination questions or tasks have a priority to differentiate children - a secondary concern is that it covers the curriculum content.
  • secondly, as an exam strategy it does not ‘pay’ for a child to help another peer in their class, because that could interfere with their overall ranking from ‘best’ to ‘worst’.
  • finally the concern for fitting the results into a ‘normal’ distribution of scores from ‘A’ to ‘E’ grade takes greater priority over the concern as to whether the children actually understand or know the curriculum content. 

That is why it’s gratifying to see the Fiji Ministry of Education moving away from this and into ‘benchmark’ testing. This is the second year that the Language and Numeracy Assessment has been administered nationally. I suspect there are rough edges to this new process, but it is early days, however the principle is far more likely to produce surgeons (– pick your profession) who actually do know their curriculum topic and content.

Education is a progressive discovery of our own ignorance.  -Will Durant