Matric results of learners involved in exam leaks to be blocked, pending investigations

After a failed attempt to have the 2020 National Senior Certificate (NSC) candidates rewrite two leaked papers, maths and physical sciences, the National Investigations Task Team (Nitt), operating under the auspices of the National Examinations Irregularities Committee (NEIC), found that the full extent of the leaks can’t be determined. 

“Given the fact that the spread of the questions papers was done via WhatsApp, it is a fact that the full extent of the leaks may never be revealed,” said Hugh Amoore, chair of Nitt. 

“But, based on the available evidence we have at the moment from the investigative marking, interviews and statistical analysis, widespread leaks did not occur.”

This is despite the department’s earlier iterations that the leaks were widespread. 

Amoore said there is evidence that 236 mathematics candidates and 62 physical science candidates were sent screenshots of questions or question papers before the exam dates of both papers. 

Nitt followed a multipronged investigation method that involved investigative marking and verification, statistical analysis of raw marks, interviews with candidates, an investigation by a private forensic IT investigator, and investigation to establish the source by audit and the Hawks. 

Results from the investigative marking showed no evidence of candidates colluding and most of the scripts showed individual work. 

Amoore said on Sunday that scores of candidates in one of the groups where the papers were leaked (a high-achievers group) were high. These learners had access to questions 6.2. and 6.4 of maths Paper 2. 

TimesLIVE reported that most of the matric pupils found to have had access to maths Paper 2 before it was written were top maths pupils from across SA who had been tutored by Stellenbosch University.

In an analysis of the chat groups no evidence could be found to suggest that these candidates benefited from prior access to the question paper or sub-questions, that candidates suspected to be directly involved in the leak generally received different marks per question, and specific candidates presented evidence of performance that needs further investigation, but this was not group-related.

One source of the leak was traced to someone who works for a Johannesburg company that was contracted by the department to print the 2020 exam papers, and the other was from the Government Printing Works. 

Recommendations made to Umalusi

In the lead-up to the education overseeing entity Umalusi verifying the matric results, Nitt made recommendations on how to handle implicated candidates. 

As per the department’s recommendations, implicated candidates were categorised into:

A.. Where the candidate was a recipient, and probably an unwitting recipient, of the WhatsApp message concerned. There are 236 such candidates.

B. Where the statistical evidence suggests that performance in P2 relative to P1 makes him or her an outlier. There are 205 such candidates.

C. Where a candidate is implicated in that he/she was sent a question/question paper by WhatsApp and the statistical evidence suggest that performance in P2 relative to P1 makes him or her an outlier.

D. Where there is further evidence (eg, that the candidate was an “initiator” or a “spreader”).

E. Where the school mean shows that the school is an outlier.

The results of candidates in category E and D will be blocked pending further investigations. Those in category A and B will receive their results pending further investigations, while category C candidate results will be released pending the school-level investigation. 

Under the General and Further Education and Training Quality Assurance Act, the founding legislation for Umalusi, the council can still cancel a certificate issued to a learner if on investigation the council finds that the learner has not met qualification requirements.

Umalusi will hold a briefing on Monday, 15 February, to announce the approval of the exams. DM