Now that the dust has settled on the matric results celebrations and the school year is in full swing, we need to take a harder look at whether high repetition rates in Grades 10 and 11 are justified, and whether we place too much emphasis on matric results.

This the key message Dr Rebecca Selkirk is delivering to policymakers. Selkirk recently obtained her doctorate in economics at Stellenbosch University, exploring grade repetition as well as matric results. Selkirk points out that before the pandemic, approximately 1.5 million learners repeated each year across Grades 1 to 12 (roughly 12% of total enrolment) with approximately one-third of all repetition happening in the lead-up to matric (Grades 10 and 11). 

Almost a third (31%) of Grade 10 learners and nearly a quarter (24%) of Grade 11 learners repeated at the end of 2019.

“We spend a lot of money on retaining learners in Grades 10 and 11, where repetition’s effectiveness is unclear, and the associated drop-out risk is highest. But despite this, repetition in these grades has received little research or policy attention,” says Selkirk.

As part of her doctoral study, Selkirk explored the link between repetition rates in Grade 11 and matric pass rates in Western Cape schools before and during the pandemic (2018-2021). She found a positive link between high Grade 11 repetition rates and stronger matric results, which she says could indicate that “schools may benefit (through higher matric pass rates) by ‘gatekeeping’ access to matric – and/or that repeating Grade 11 helps learners catch up and improve their chances of passing matric.” 

However, she notes that “interestingly, when repetition rates in the province dropped substantially at the end of 2020 (from 12% to 6%) due to pandemic-related leniency, the positive link became weaker in both size and significance”.

Selkirk also estimated what the Western Cape’s matric results in 2020 and 2021 would have looked like in the absence of the pandemic, assuming that the 2018 and 2019 returns had still applied to the much lower Grade 11 repetition levels. 

“Estimates showed that schools’ achievement in 2021 was much higher than expected given the lower repetition rates, particular in less resourced (Quintile 1 to 4) schools where the changes in repetition patterns were most stark, and learners would have been expected to be more exposed to the negative impacts of the pandemic.” 

According to Selkirk, this raised questions about whether lower repetition rates may have unintentionally inflated the 2021 matric results by influencing the mark standardisation process. “If grade inflation was not the cause, then it also prompts another question: should Grade 11 repetition rates be allowed to return to pre-pandemic levels, if the ‘gains’ schools received from maintaining high repetition rates were artificial and only applicable in a system where repetition was widespread?”

She says national matric achievement was also, at face value, better than expected despite even larger declines in repetition (national Grade 11 repetition rates dropped from 24% to 11%). “After an initial decline and stabilisation in 2020 and 2021, the matric pass and bachelor pass rates broke records by 2023 and 2024. And in absolute numbers, matric success over the 2020 to 2024 period was staggering: these five years produced the pre-pandemic equivalent of six and a half full years’ of matric passes, while Bachelor passes amounted to more than eight years’ worth.”

Given these surprising results, Selkirk investigated national matric results to explore if observable learner-level characteristics such as overage status (proxying for repetition), subject choice, and relative academic ability could explain matric performance and its changes between 2011, 2019, 2021, and 2023. She found that, despite large shifts in subject choice and matric age profiles over time, little of the overall improvement in results by 2023 could be explained by these and other factors included in her models.

In interpreting these results, Selkirk cautions that while unintentional grade inflation may have played a role in higher-than-expected matric achievement, it is unlikely that this would be the sole cause – particularly in explaining the improvements after 2021, where ‘shocks’ to the repetition rates were not as large as those experienced at the end of 2020. 

“It is possible that pandemic-related support programmes more than compensated for any learning losses, and that this additional support was more beneficial than repetition for learners who would normally have repeated Grade 10 and/or 11 pre-pandemic.”

She goes on to explain that “it is also possible that lower repetition on its own was beneficial (for example in reducing class sizes, avoiding negative psychosocial effects such as academic disengagement for would-be repeaters)—2023 had the highest-ever proportion of matric candidates who were the correct age-for-grade, a factor which is usually positively associated with achievement”. 

She points to recent evidence that shows Grade 10 and 11 repetition rates returning to the high levels observed before the pandemic, suggesting that high repetition will remain without policy intervention. 

“There is enough evidence to suggest that our high Grade 10 and 11 repetition rates may not be justified, and that alternatives to repetition could be equally or more beneficial to struggling learners. More research is urgently needed in this area because the potential cost savings could free up resources for improvements elsewhere in the education system”. 

Selkirk also calls for reduced overall emphasis on matric pass rates, noting that “if we place too much emphasis on matric pass rates, we risk incentivising schools to make decisions—like maintaining these high repetition rates, encouraging ‘easier’ subject choices, focussing resources on matric alone—that may at best be sub-optimal and, at worst, harmful to learners or the education system as a whole.”