New figures show about a third of NCEA school marking examined by official moderators last year did not meet national standards.
The New Zealand Qualifications Authority data showed 28 per cent of moderators' judgments on teachers' marking last year recommended the teachers' work was reviewed. The students' work
could have been marked either too easily or too strictly.
The majority of NCEA credits students gained last year were through internal assessments.
Authority deputy chief executive (qualifications) Bali Haque said the check marking rate was "reasonable" as schools were asked to submit work they considered to be at the borderline. The rate recommended for review has stayed largely unchanged since 2004.
But Mr Haque said the policing of internal assessment gets a boost this year as authorities move to better motivate students and lift public confidence in the system.
"This year the system is shifting and we'll be keen to look at the impact that has on the numbers this time next year."
The amount of moderation was set to be tripled this year and random sampling introduced, as opposed to schools being able to select the work being policed.
"The boundary work is actually quite important for professional development for teachers," he said.
"But from [Weekend Herald] readers' perspective, if they want to know whether we are going to do real checks - real random samples - the answer is 'yes'."
Mr Haque said more real-life examples of students' work showing how it was marked and why would be made available on the internet to guide teachers. He did not have specific figures but said the plan was to considerably build up the number of "exemplars", as they are known.
This week, the NZQA released its school-by-school breakdown of NCEA pass rates. While comparing the schools based solely on the data is difficult, Mr Haque said the statistics provided useful information for schools and teachers.
