Option 1: Focus the Curriculum Tests on core subjects to prioritise time and energy for teaching and learning.
Option 2: Use Curriculum Tests insights to identify and scale support for learning gaps and weaknesses.
Option 3: Maintain the vision of the GEC as a holistic report card.
The learning outcomes in South Africa and the modest improvements over the past decade are of central importance for the country’s education system and its society. A well-evidenced lever for educational improvement is valid and reliable assessment which enables teachers to accurately understand learning outcomes and adjust pedagogy in response (OECD, 2013[33]). School-based assessment and locally‑developed exams play a central role in South Africa’s assessment system, particularly in the absence of standardised assessment across the full cohort (prior to the National Senior Certificate [NSC] in Grade 12). South Africa has developed systems for guiding and structuring assessment at the school level to encourage standardisation, but such an approach cannot remove all potential variations for how learners are assessed against national learning expectations. The full cohort, standardised Curriculum Tests introduced by the General Education Certification (GEC) are an important opportunity to assess all learners against common national standards for literacy and numeracy. This section discusses the GEC’s potential to play a critical role in South Africa’s education system and how it can be fully utilised to promote high standards for all learners.
Prior to the introduction of the GEC, school-based assessment activities and school-set exams1 constituted all the marks that make up a learner’s end-of-year grade in every year of school except the final year, Grade 12 (at which point NSC grades are determined 25% by SBA and 75% by final examinations). In Grades 7-9, 60% of learners’ end-of-year grades are based on SBA throughout the year and 40% is based on examinations at the end of the year, which are also typically set at the school level (DBE, 2023[34]).
While many systems around the world use school-based assessment and school-set exams as the main form of assessment prior to the end of upper secondary education, a key difference between most systems and South Africa are the attached stakes. End-of-year grades, or promotion grades in South Africa carry significant stakes for learners since they determine promotion to the next grade. Grade repetition is a relatively frequent practice in South Africa (see Context section).
Assessment set and marked internally by the school – like SBA in South Africa – provides teachers with significant flexibility around the tasks, conditions and content of the assessment. The flexibility of internal assessment – such as when assessments are set and the tasks used – can support integration with teaching and learning providing formative value for learners who can use results and teacher feedback to deepen and strengthen their learning (McDowell, Sambell and Davison, 2009[35]). However, wide variations in teachers’ assessment practice across schools and classrooms can undermine the reliability of school‑based assessment (OECD, 2026[36]).
In South Africa, the CAPS provide structure and prescription to promote some consistency in how teachers use assessment. CAPS guidance sets out the number and type of assessment tasks teachers are expected to use by grade and subject. For example, in English as a Home Language in Grade 9, teachers are required to use ten assessment tasks throughout the year, including four oral tasks such as prepared and unprepared reading, group discussion and role plays; three writing tasks such as narrative, descriptive and argumentative essays, letter writing and dialogues; and three tests or exams (DBE, 2011[37]). The degree of prescription encourages SBA to be used as intended, so that teachers in South Africa draw on a variety of different assessment tasks while maintaining a degree of commonality across different schools and classrooms.
There are well-documented concerns in international research about the reliability of school-based assessment related to the consistency and quality of marking and judgement at the school level, and alignment with agreed national standards or curricular outcomes (OECD, 2026[36]). Evidence from different education systems demonstrates how far teacher grades may diverge from expected national learning standards. For example, a national investigation in Sweden has recently concluded that, despite multiple national initiatives to promote greater comparability in teachers’ grades, grades from teacher-based assessment do not represent the same level of knowledge and skills across schools (Skolverket, 2020[38]). While some differences between external examinations and teacher-based assessment are natural, as they assess different competencies, consistent divergences across systems underscore the inherent challenges of teacher-based assessment for ensuring that all learners are progressing in line with national expectations.
To promote the quality of SBA tasks and the reliability of assessment judgements, South Africa has a developed system of quality control and social moderation at school, district and province level. For all class tests, teachers engage in collective, social moderation by sampling several scripts to discuss the marking to promote standardised marking. A sample of marked scripts are then moderated by the school’s moderator. Schools keep a record of their assessment tasks and internal moderation reports which may be checked at the district and province level (DBE, 2023[34]).
Despite the significant investment in social moderation in South Africa, evidence suggests that there are wide variations between SBA marking and national standards. Since 2015, the Western Cape has been the only province to use full cohort, annual learner assessments. The data from the standardised learner assessments in the Western Cape suggests that there are significant variations between SBA and standardised assessment results. For example, in 2022 in Grade 3 in mathematics, the SBA score distribution was concentrated around 60‑70% with a peak near 65%. In contrast, systemic test scores based on the same learning standards (i.e. the Western Cape’s standardised assessment) were concentrated at lower levels, with most scores between about 35% and 50% and a peak around 40%. The contrast between results suggests that while most learners were not reaching the required standard, their teachers believed that they were. (World Bank, 2024[39]). The findings from Grade 3 in the Western Cape are mirrored by similar research based on SBA in Grades 11-12 and examination results from the NSC in Grade 12. This research has found significant differences between SBA and the external NSC marks, with school-based marks often being consistently higher than NSC marks (van der Berg and Shepherd, 2009[40]). Research also suggests that disparities between SBAs and external exams do not occur evenly across schools, with more socially disadvantaged schools demonstrating greater disparities (where school-based marks exceed external exam results by greater amounts) on average (van de Berg et al., 2023[41]; Euston-Brown, 2025[42]).
Given the complexities associated with school-based assessment, most OECD countries implement national assessments (28 out of 32 countries, for lower secondary education) to provide teachers – and the system as a whole – with reliable assessment data (see Figure A B.1). Among the countries with national assessment, the majority (18 out of 28) organise them annually across the full cohort. In cases where national assessments are only sat by a sample of learners, these national assessments are frequently organised annually (5 out of 10). By sampling the full cohort and/or implementing national assessments annually, these countries have accurate, representative data that they can use to closely monitor trends and understand the strengths and challenges of teaching and learning in their system.
Teachers and schools in South Africa report providing support for learners to address weaknesses in their learning. In TALIS 2024, almost all teachers in South Africa (92.6%) report frequently or always adapting teaching methods to learners’ needs (OECD average 84.1%) (OECD, 2025[7]). However, this provision may not be sufficient or address the deeper needs of struggling learners. Learners indicated to the OECD team that they do not always feel comfortable revealing what they do not know or understand to their teachers. Schools do commonly provide catch-up classes before and after school; however, attending before and after school classes also requires motivation and confidence, which learners with lower skills might lack.
One of the other main tools to address learning gaps in South Africa is grade repetition. While there is some evidence that repetition can have some learning benefits in earlier years, these benefits tend to be short term (Wills, 2023[43]). In secondary education, repetition is found to have long-term negative effects on performance and completion (Valbuena et al., 2020[44]). Repetition also has negative effects for South Africa’s economy and public spending; it is estimated to absorb between 8-12% of public education spending (van der Berg et al., 2019[45]). Estimates suggest that reducing repetition could result in major savings – as much as 2 billion South African Rand annually (almost EUR 10 million in 2025) if reduced by around 3% across grades (World Bank, 2025[24]). Reducing repetition could also release resources that could be directed elsewhere, such as to targeted, well-evidenced measures for improving low skills.
The centrally set Curriculum Tests introduced by the GEC can play an important role in assessing every learner against common national standards. Standardised assessment instruments, like the Curriculum Tests, can efficiently provide reliable insights into learner progress, strengths and fragilities – enabling teachers to improve their practice and target support to learners who need it the most. However, the current load of Curriculum Tests within the GEC is comparatively heavy. There are externally developed tests in five subjects – First and Home Languages, Natural Sciences, Economic and Management Sciences and Mathematics. Schools also develop end-of-year tests in the remaining four subjects – Technology, Social Sciences, Creative Arts and Life Skills. At the system level, central co-ordination is required to develop the five centrally set Curriculum Tests, and high levels of quality assurance are needed to ensure credibility and validity. For schools and teachers, developing new Curriculum Tests in the remaining four subjects adds new burdens, alongside existing workload burdens associated with SBA.
To ensure that the time and resources required to develop, administer and mark the new Curriculum Tests is manageable, South Africa could prioritise centrally-developed Curriculum Tests in maths and language – and not develop centrally-set Curriculum Tests in the other subjects nor require schools to develop additional Curriculum Tests in the other remaining subjects. Focusing the Curriculum Tests on maths and language is efficient since these two subjects underpin learning across many other subjects and successful functioning at work, social and personal life (Graham, Kiuhara and MacKay, 2020[46]; Thompson, Bowling and Markle, 2018[47]). Focusing the external Curriculum Tests on maths and language would help to focus national conversations on learning in these two critical domains.
Implementing end-of-year Curriculum Tests across the curriculum is relatively burdensome. Across OECD countries with available information, only around a third (13 out of 32) have national examinations in lower secondary education. Among those with national exams in lower secondary, few cover as many subjects as the Curriculum Tests in South Africa’s GEC, typically just two to four subjects (OECD, 2023[48]). In a context where teachers already report spending a large amount of time marking and where public funding for education is expected to decline in the coming years (World Bank, 2025[24]), reducing the number of Curriculum Tests to focus on the most essential subjects could help to promote efficient resource use.
The GEC Curriculum Tests in language and maths can provide schools with key information about learners’ strengths and weaknesses, as well as supporting wider systemic understanding of the skills and knowledge developed by Grade 9.
Once learners have taken the GEC, they still have another three years in school. For learners with weak skills this is a critical – and perhaps final – window for the education system to provide them with systematic support through full-time education to address skills weaknesses. To identify what measures are likely to be most effective in South Africa, the country could undertake a comprehensive review of available evidence to identify what approaches internationally, regionally and locally are proven to be effective in raising learning outcomes at the end of secondary education.
One option is to provide smaller group or one-to-one instruction targeted in maths and language for learners in need of additional support. Meta-analysis covering a wide range of programmes and locations has found that small-group and one-to-one instruction is associated with positive, large improvements in learning (Nickow, Oreopoulos and Quan, 2020[49]). South Africa might adapt smaller group instruction programmes to the national context. For example, while one-to-one tutoring may not be feasible, perhaps smaller group tutoring is. Similarly, while teachers may not have the time to provide such tutoring, paraprofessional tutors, which might include other school staff and undergraduate learners, are also found to be effective. Reducing the role of SBA in Grade 9, given the high amount of teachers’ time this takes up and the range of other educational staff it requires, could free up school staff time for smaller group tuition in Grade 10. Finally, smaller group tuition is found to be more effective when it takes place during the school day rather than after school, to ensure that it happens and that all the learners who need to access it can do so. Reducing the burden on schools to develop Curriculum Tests could also free up resources for teachers to spend more time on smaller group or one-to-one instruction.
A complementary approach would be to ensure that teachers in South Africa have appropriate materials for the teaching and learning needs of their learners. When the results from the GEC identify learning needs, this information should be systematically shared with Grade 10 teachers in language and maths to ensure that they have this information for their planning and preparation well before the start of the school year. To facilitate teachers’ use of results for adapting pedagogy, the DBE might provide schools and teachers with information about how to address common weaknesses across key content or skills areas. Guidance for teachers might draw on the evidence which identifies instructional approaches focused on the learning level, rather than learners’ grades, to be effective (Akyeampong et al., 2023[50]).
Given the central importance of raising learning outcomes in South Africa, the country might reconsider how and where teachers spend their time in assessment activities throughout school. Teaching staff currently devote a lot of time to marking learner work: according to TALIS 2024, South Africa’s teachers spend more time than any other TALIS-participating country marking and correcting students’ work. (OECD, 2025[7]). Despite the time consumed by marking and moderation, learner marks for SBA vary significantly across schools and in line with national standards. The country could consider reducing the amount of formal or reported SBA in Grades 1-8 to free up teacher time which might instead be focused on supporting struggling learners. To ensure that all learners are progressing in line with expectations and identify those who need support, schools can be encouraged to use centrally developed diagnostic tests at the start of the school year [see also (OECD, 2026[51])]. The information from such assessments would provide a reliable and valid measure of learning to enable teachers to adjust to learner needs. Implementing tests at the start of the school year can help to ensure that they are focused on diagnosing student learning rather than being associated with system-wide stakes. For example, France has recently introduced national assessments which now occur at the start of every grade in primary education, and in selected years in secondary education (Ministère de l'Éducation nationale, 2025[52]).
In many OECD countries – as in South Africa – most learners remain in education until at least upper secondary education. As such, a formal certificate at the end of lower secondary that communicates what a learner knows and can do – with final grades for each subject – to external stakeholders beyond the education system (i.e. employers) has little relevance. Only a minority of OECD countries award certificates at the end of lower secondary. Where these certificates do still exist, they tend to have evolved into a broader reflection of a young person’s knowledge and skills as opposed to representing an exit point or being a requirement for further progression. In France, for example, the diplôme national du brevet (“brevet”) is an attestation of a young person’s knowledge and skills at the end of lower secondary (Ministère de l'Education Nationale et de la Jeunesse, 2024[53]).
In South Africa, the GEC was initially planned as a new certificate at the end of compulsory education. Yet introducing new certification at the end of lower secondary risks hindering progression to upper secondary if it becomes a mandatory requirement for accessing Grades 10-12. Given the awarding of the GEC would coincide with the end of compulsory education, this could also create the perception that the GEC is a suitable exit qualification for learners in South Africa. Confusion over the purpose of the GEC – whether it is an exit qualification or a holistic recognition of learners’ skills ahead of transitions to upper secondary – may be compounded by the fact that the GEC is being proposed as an exit qualification for one specific cohort of learners: those attending Schools of Skills.
The GEC is intended to recognise learners’ broad set of skills and knowledge and support them to identify their pathways into the final stage of school and post-school. To best reflect this original intention of the GEC, a report card rather than a formal certificate could be used to communicate learner results. The report card might set out learners’ results separately for the Integrated Project, Curriculum Tests, learners’ self-evaluation of their 21st Century Skills (see Policy Issue 2), and other qualitative information such as highlights of learner achievement and progress and Inclinations Assessment results. Specifically, this would mean:
Reducing the assessment load of SBA within the GEC:
Currently, the GEC communicates final grades for each subject, with 60% based on SBA (including Term 1 and Term 2 SBA and Term 3 Integrated Project results) and 40% based on Curriculum Tests. This approach adds unnecessary stakes to Term 1 and Term 2 SBA, diverting energy from disciplinary foundations to assessment preparation and task completion. Not including SBA results from Terms 1-2 in the GEC would provide space for more teaching and learning of formative value which is an essential function of SBA (OECD, 2005[54]).
Report results from the Integrated Project separately
At present, learners’ results from the new Integrated Project are split into individual subjects and combined with SBA results. This restricts the potential of the Integrated Project to signal learners' acquisition of cross-cutting competencies like communication and collaboration. Separately reporting the results from the Integrated Project could also help to elevate the visibility of this innovative assessment as part of the GEC and the education system overall.
Report Curriculum Tests results in language and maths separately
Language and Maths Curriculum Test results should be prominently displayed on the GEC report card to help teachers quickly identify learners with weak skills in these foundational areas and promote good use of new reliable and valid assessment data in the country.
The GEC could potentially act as an exit certificate for learners in Schools of Skills (which serve learners with Special Education Needs) who might achieve its learning objectives at older ages. To provide these learners with stronger foundations for transitions to employment, South Africa is also exploring how to recognise learners’ vocational skills via other vocational qualifications and short courses.
There are many complexities related to how the achievements of learners with SEN are certified within national certification frameworks, including the heterogeneity of learner skills, knowledge and understanding, effectively reflecting a wide range of learner achievement, and interactions with other certificates and future pathways. Some systems, including British Columbia (Canada) and Ireland offer dedicated certificates for learners with SEN. For example, the School Completion (“Evergreen”) Certificate in British Columbia (Canada) is used to recognise the accomplishments of learners with disabilities and is based on achievement of their Individual Education Plan (British Columbia Government, 2025[55]). Dedicated certificates for learners with SEN can be a powerful goal for learners to work towards. It is beyond the scope of this Policy Perspective to consider how to best meet the needs of the Schools of Skills in assessment and certification. However, careful reflection will be needed to determine whether a certificate like the GEC – which also aims to meet the needs of the entire learner population – best meets the needs of learners in the Schools of Skills.
← 1. Some might be developed at the provincial or district level.