Education body defends method of standardising exam results despite criticism

UMALUSI, the council for quality assurance in general and further education and training has defended its standardisation process following criticism including claims that the process inflated matric results.

Since December, the council has been criticised for the 2016 National Senior Certificate results, with some claiming the process was not implemented in a consistent manner.

In an open letter to the council, DA MP Gavin Davies said the council had standardised marks in cases where results were lower than the historic norm but in cases where the results were higher, they had not been brought down to the norm.

This, Davies said in his open letter to the council chief executive, Mafu Rakometsi, showed inconsistency and a tendency to only adjust marks upwards.

In a bid to unveil the standardisation process, Rakometsi told journalists this week that by law, the council had the mandate to adjust raw marks according to set standards following the historic norm of pupil performance per subject over a period of five years.

He said the council had an assessment standards committee made up of independent education experts responsible for the process of standardisation.

According to Rakometsi, the process was scientific and was aimed at mitigating the effect of factors other than a pupil’s knowledge and aptitude on their performance.

The process took 18 months and involved moderation of question papers, review of pupil performance against historical performance of candidates in each subject, pairs analysis of related subjects and statistical moderation of school-based assessment, exam readiness, marking guidelines and mark verification, among other things.

As part of his concerns, Davies said the standardisation process tended to adjust marks upwards, thus painting a picture of results better than they actually were.

Rakometsi said this was not the case. Marks were adjusted up and down to be brought in line with the median of historical performance.

He said in cases where raw marks showed a considerable change from the norm, that warranted questions about whether the paper was pitched at a level higher than the pupils’ ability or on the other hand was too easy.

“If you pose an examination question asking on which date is Christmas Day, all the pupils are likely to get that question right,” he said.

In the same way, there were questions that almost all pupils got wrong, Rakometsi said.

He said they assessed what could have caused a major difference in marks, including the wording and how pupils were likely to understand a question.

Three-quarters of the total mark, the written component, was subject to variation, while the remaining quarter came from a pupil’s school assessment.

noloyiso.mtembu@inl.co.za