OCDE - Developing South Africa's National Student Assessment Framework

 1

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

6 July 2026 South Africa has made significant progress in expanding access to education over the past three decades. However, low learning outcomes and persistent inequalities between schools rem ain key challenges. Recognising the role of assessment in improving education quali ty, South Africa’s Department of Basic Education requested t he OECD’s advice on how to further develop its National Student Assessment Framework. This Policy Brief summarises the conclusions of this analysis, and puts forward the following policy recommendations: • Update the National Student Assessment Framework to include all major assessments and clarify their key characteristics, including their purpose and use of results . Including the General Education Certificate and the National Senior Certificate will be important as they represent key milestones in students’ progression. The Framework should provide direction on South Africa’s school-based assessment policy to better balance formative and summative functions. This may require reconsidering how assessments contribute to term or endof-year marks. Finally, explicit mentioning of key objectives and indicators from the Strategic Plan 2025–30 would promote greater policy coherence and enhance the Framework’s role in system-level monitoring and policy making. • Expand the National Student Assessment Framework to include additional standardised assessments resources. To address gaps in reliable assessment data, South Africa should consider introducing additional standardised assessment resources, such as nationa l diagnostic assessments and an item bank, to support teaching and learning. Administering national diagnostic assessments at the start of the school year or term would help maintain their diagnostic focus and support a more balanced approach between formative and summative school-based assessment . • Use the National Student Assessment Framework to guide the further development and implementation of South Africa’s Systemic Evaluation. While the Systemic Evaluation is welldesigned, the level of detail and content of the proficiency-level descriptors vary across grades and could be improved. In mapping the assessment framework against international benchmarks, the OECD also identified a couple of areas where the content of the tests might be further developed.

implementation of South Africa’s Systemic Evaluation. While the Systemic Evaluation is welldesigned, the level of detail and content of the proficiency-level descriptors vary across grades and could be improved. In mapping the assessment framework against international benchmarks, the OECD also identified a couple of areas where the content of the tests might be further developed. Any changes should align with the revised school curriculum. South Africa might also consider piloting digital assessment in future Systemic Evaluation cycles. This coul d shorten reporting timelines and offer additional benefits, such as strengthened data analytics. • Further enhance the monitoring of teaching practice and teachers’ working conditions through the National Student Assessment Framework . The Systemic Evaluation Teacher Questionnaire could be strengthened in several areas, such as the monitoring of teachers’ workload and work-related stress, their professional development, and their pedagogical and assessment practices, and prioritised in line with national objectives. Field-testing the r evised questionnaire would help avoid significant increases in survey length. The Framework sho uld also provide direction on future participation in the Teaching and Learning International Survey (TALIS ) and set out plans for analysing and sharing data on teachers. Developing South Africa’s National Student Assessment Framework 2

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

Why is South Africa in updating its National Student Assessment Framework? South Africa has expanded access to education over the past three decades. However, low lear ning outcomes and wide inequalities between schools remain a challenge. Results from the country’s new national standardised student assessment, South Africa ’s Systemic Evaluation ( SASE), reveal that large shares of students in Grades 3, 6 and 9 perform at the lowest proficiency levels in reading and numeracy. In Grade 3, for example, 37% of students in 2021 were “Emerging” level readers . This means they were just beginning to develop grade-level foundational knowledge and skills and require extensive guidance and support. International assessments confirm these patterns (see Figure 1) an d help explain the high

In Grade 3, for example, 37% of students in 2021 were “Emerging” level readers . This means they were just beginning to develop grade-level foundational knowledge and skills and require extensive guidance and support. International assessments confirm these patterns (see Figure 1) an d help explain the high rates of grade repetition and dropout later in schooling. These outcomes reflect long-s tanding structural inequalities among South African schools, including uneven access to qual ified teachers and learning resources. In this co ntext, South Africa’s Department of Basic Education has made raising learning outcomes a core objective in its Strategic Plan 2025-30 (Department of Basic Education, 2025[1]). Figure 1. Low learning outcomes are a key challenge in South Africa

Note: Grade repetition rate calculated based on the number of students who will repeat in the next year, divided by the total students in the current year.

Source: A : von Daviet et al. (2024[2]), TIMSS 2023 International Results in Mathematics and Science,

https://doi.org/10.6017/lse.tpisc.timss.rs6460; Mullis et al. (2023[3]), PIRLS 2021 International Results in Reading, https://doi.org/10.6017/lse.tpisc.tr2103.kb5342; B : Department of Basic Education (2024[4]), South African Systemic Evaluation 2022 Volume 2 Learner Proficiency Report. Recognising the role assessments can play in helping to improve learning outcomes, the Department of Basic Education is now further developing the National Student Assessment Fram ework. This framework will set out South Africa’s overarching vision for student assessment and intends to define the primary purposes and key characteristics of different assessments (e.g. classroom-bas ed assessments, examinations, and national standardised assessments). It will also explain how di fferent assessments complement each other to support quality objectives across the school system, such as identifying learning gaps and providing targeted interventions to support student learning.

250 275 300 325 350 375 400 425 450

examinations, and national standardised assessments). It will also explain how di fferent assessments complement each other to support quality objectives across the school system, such as identifying learning gaps and providing targeted interventions to support student learning.

250 275 300 325 350 375 400 425 450 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 202 3

Panel A: Student performance on TIMSS and PIRLS over time

Mathematics Grade 8 Science Grade 8 Reading Grade 4 International Low Benchmark 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Grade 3 Reading Grade 3 Numeracy Grade 6 Reading Grade 6 Mathematics Grade 9 Reading Grade 9

Mathematics Panel B: Student performance on South Africa's Systemic Evaluation 2021

Emerging Evolving Enhancing Extending 3

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

Features of national student assessment frameworks internationally While national student assessment frameworks vary in form and function across c ountries, they have common attributes (OECD, 2013 [5]). Typically presented as a policy document, national student assessment frameworks clarify the purposes of different assessments and explain how they relate to each other and national education objectives . Well-designed frameworks can strengthen accountability and public confidence in the school system by transparently reporting results. By aligning assessment, curriculum and system objectives , they can also strengthen teaching and learning and help avoid duplication and over-testing. Key features of well-designed national student as sessment frameworks include: • Clear statements of purpose that explain how assessments will support student learning and can be used to orient decisions on test design and the use and reporting of test resul ts. For example, assessments used for system-level monitoring imply different choices in terms of test design, sample size and frequency than those used for diagnosing individual student learning;

be used to orient decisions on test design and the use and reporting of test resul ts. For example, assessments used for system-level monitoring imply different choices in terms of test design, sample size and frequency than those used for diagnosing individual student learning; • An overview of how assessment supports other aspirations for the school sy stem – notably how assessments will promote the learning and pedagogical goals of the curriculum and whe ther and how assessments will be used for monitoring and accountability at different levels of th e system; • A strong emphasis on ensuring coherence and balance in the use of assessments – including attention to the overall volume of assessments and duplication, and how potential tensions between the formative and summative functions will be managed. Features of South Africa’s draft National Student Assessment Framework In 2022, the Department of Basic Education developed a Concept Document outlinin g an initial draft of South Africa’s National Student Assessment Framework (Department of Basic Education, Unpublished[6]). The Concept Document explains that the Framework intends to serve as a “basket of fit for purpose assessment programmes targeted at measuring and monitoring learning outcomes and supporting teachers in assessment practice” (ibid.). It outlines three interrelated pillars of formative, summative and system-wide wide assessments (i.e. national and international assessments). It also mentions the different assessment instruments currently in use in South Africa (see Figure 2). T he Department subsequently developed a detailed PowerPoint presentation to reflect its evolving vision for the National Student Assessment Framework. Figure 2. Overview of South Africa’s national assessment and survey system

Source: Developed by the authors based on the draft NAF Concept Document and supporting materials. 4

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

While the Concept Document and PowerPoint presentation have informed policy dia logue with education stakeholders, they have not been publicly released . Moreover, it is not clear how data will be used to monitor progress towards national objectives for learning set out in the Department of Basic Education Strategic Plan 2025-30. There are also some notable gaps in these supporting material s. For example,

stakeholders, they have not been publicly released . Moreover, it is not clear how data will be used to monitor progress towards national objectives for learning set out in the Department of Basic Education Strategic Plan 2025-30. There are also some notable gaps in these supporting material s. For example, while the key characteristics of SASE are well defined (except for its frequency) , there is a lack of information on school-based assessments and national examinations. Witho ut further developing and publicly releasing the National Student Assessment Framework, it will be d ifficult for South Africa to promote a coherent set of assessment programmes to monitor student performance, strengthen teaching and learning and build public confidence in the school system. National standardised assessments The Concept Document of South Africa’s National Student Assessment Framework references two national standardised assessments: SASE and the Early Learning National Ass essment (ELNA) . The SASE, which was introduced in 2021, is a sample-based national assessment of reading and mathematics for students in Grades 3, 6 and 9. While not clearly stated in the Concept Document, South Africa plans to administer the SASE every three years, with the next cycle scheduled for 2026. The Concept Document explains that the SASE is designed to provide comparable data across grades and provinces to inform policy and monitor progress over time. The OECD conducted a comparative mapping of the SASE assessment framework against those of international reference assessments. This analysis found SASE’s framework to be well designed for its monitoring purpose and broadly aligned with the domains, test design features and reporting processes of international assessments. However, differences in content coverage and proficiency levels were also identified as warranting potential modifications to the SASE framework in the future (see Box 1 and Recommendation 3: Use the National Student Assessment Framework to guide the further development and implementation of South Africa’s Systemic Evaluation ). ELNA is a sample-based national assessment of Grade 1 students that was admini stered in 2022 and

2025. ELNA aims to measure the emerging literacy and numeracy skills of Grade 1 students as they enter formal schooling and to assess their readiness for learning. The assessment generate s national and

ELNA is a sample-based national assessment of Grade 1 students that was admini stered in 2022 and

2025. ELNA aims to measure the emerging literacy and numeracy skills of Grade 1 students as they enter formal schooling and to assess their readiness for learning. The assessment generate s national and provincial estimates of students’ early literacy and numeracy levels, and provides insi ghts that can inform targeted support and interventions. However, the Concept Document provides limited information on ELNA. It briefly identifies the target population (Grade 1 students) and the domains as sessed – emerging literacy, numeracy, and school readiness skills – but other key details, such as its purpose (outlined above) and frequency, are not provided.

International assessments and surveys The Concept Document and supporting material state that South Africa will continue to participate in the Trends in International Mathematics and Science Study (TIMMS) , Progress in International Reading Literacy Study (PIRLS), the Southern and Eastern Africa Consortium for Monitoring Educational Quality (SACMEQ) and OECD’s Teaching and Learning Survey (TALIS) to benchmark performance internationally and inform education policy. However, reference to the country’s participation in TALIS, which collects data from teachers and school leaders on teaching practices, professional development and school leadership, was not included in the 2022 Concept Document, only in the supporting material. 5

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

Box 1. Main findings from the OECD mapping of the SASE assessment framework against international assessments The OECD’s mapping of the SASE assessment framework against selected international reference assessments – PISA, PIRLS, TIMMS and the Southeast Asia Primary Learning Metrics (SEA-PLM) – focused on: assessment domains (including content strands, cognitive processes and contexts); test design features (such as item proportions and item types); and approaches to reporting results (including proficiency scale scores and levels). It revealed: • SASE is a well-designed national assessment for system-level monitoring of student performance . It provides a

content strands, cognitive processes and contexts); test design features (such as item proportions and item types); and approaches to reporting results (including proficiency scale scores and levels). It revealed: • SASE is a well-designed national assessment for system-level monitoring of student performance . It provides a comprehensive picture of student learning outcomes at key stages, tracks national and provincial trends, and generates evidence to inform policy and programming. Questionnaires for students, parents, school principals, teachers, a sample of district officials and a separate Whole-School Evaluation Questionnaire provide valuable contextual information to consider alongside results. • The SASE assessment framework is clear, but some aspects could be modified. While the SASE framework covers similar content domains and cognitive processes as international assessments, some differences were identified. In SASE’s mathematics framework, the coverage of Data (see Table 1. Comparison of the content strands in mathematics frameworks ) is relatively lower and proficiency levels place less of an emphasis on broader mathematical competencies related to problem solving. In reading, there is broad equivalence between SASE’s reading comprehension and the reading literacy frameworks of PISA, PIRLS, and SEA-PLM, but, unlike the latter, SASE does not cover writing skills. Table 1. Comparison of the content strands in mathematics frameworks SASE PISA (15-year-olds) TIMSS/PIRLS SEA-PLM (Grade 5) Grade 4 Grade 8 Numbers, Operations and Relationships: (G3:50; G6:40; G9:20) Change and Relationships (25) Number (50) Number (30) Number and Algebra (35-45) Patterns, Functions and

Algebra: (G3:10; G6:10; G9:30)

Algebra (30) Space and Shape

Geometry: (G3:15; G6:20; G9:20)

Space and Shape (25) Measurement and Geometry (30) Measurement and Geometry (20) Measurement and Geometry (35-45)

Measurement:

Space and Shape

Geometry: (G3:15; G6:20; G9:20)

Space and Shape (25) Measurement and Geometry (30) Measurement and Geometry (20) Measurement and Geometry (35-45)

Measurement: (G3:15; G6:20; G9:20) Quantity (25)

Data Handling: (G3: 10; G6:10; G9:15) Uncertainty and Data (25) Data (20) Data and

Probability (20) Chance and Data (15-25) Note: Numbers in parentheses refer to intended percentage of test items that correspond to each domain dimension. The column for the Systemic Evaluation shows the respective percentages for Grades 3, 6 and 9. • Reporting is comprehensive but timeliness and clarity could be improved. The analysis and reporting of SASE results target key stakeholders and is of high technical quality. However, there is a notable time lag (close to two years) between the SASE’s administration and when its results are available. Moreover, the level of detail and content in the proficiency-level descriptors vary across grades. In mathematics, the Grade 3 and Grade 6 proficiency levels emphasise progression in general cognitive skills (e.g. “estimate and measure”), whereas the Grade 9 descriptors provide a more balanced combination of cognitive processes and concrete skill expectations. In reading, the Grade 6 proficiency levels offer more detailed performance descriptions than those of Grade 3, while the Grade 9 descriptors are substantially less detailed.

Source: OECD (Unpublished [7]), Mapping South Africa's System Evaluation Assessment Framework against international refer ence assessments - PISA, PIRLS, TIMMS and SEA-PLM. 6

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

School-based assessments School-based assessment (SBA) – assessments developed and administered by teachers at the school level – plays a central role in South Africa’s assessment system, particularly in th e absence of full-cohort

work; more than any other TALIS-participating country (OECD, 2025 [8]). Diagnostic assessments The Concept Document outlines plans to develop diagnostic assessments starting from 2022. Since then, the Department of Basic Education has developed diagnostic assessments for Grades 3, 6, and 9 to identify gaps in literacy and numeracy skills and to inform teaching and student lea rning. The diagnostic assessments comprise test items, guidelines for teachers, and marking schemes. Their administration and marking are carried out by teachers in schools. Given low student outcomes, the limited reliability of teacher assessments, and absence of full-cohort, standardised assessment data, there is a clear need to expand access to standardised diagnostic and other formative assessments, including for additional subjects. National examinations Although the Concept Document does not explicitly refer to the National Senior Certificate (NSC), the supporting material does, indicating an intention to cover all assessments from Grades 1 to 12 in the Framework. The NSC is South Africa’s main school-leaving qualification. It is awarded at the end of Grade 12 and is aligned with Level 4 of the National Qualifications Framework. The NSC combines school-based assessment and national examinations, and its results determine access to different post-school education and training options. South Africa has also recently developed a new Grade 9 qualification alig ned with Level 1 of the National Qualifications Framework, the General Education Certificate (GEC). The GEC combi nes school-based assessment with nationally developed curriculum tests (i.e. examinations) in l anguages, sciences, economics and mathematics, alongside locally developed assessments in other subjects. It also includes an inclination (career aptitude) assessment to guide learners’ choices of up per secondary pathways (academic, vocational or occupational). Following a pilot phase, the Department of Basic Education began phased implementation of the GEC in 2025, with full roll-out expected by 2027. As with the NSC, the Concept Document does not explicitly refer to the GEC, although i t is referenced in

(academic, vocational or occupational). Following a pilot phase, the Department of Basic Education began phased implementation of the GEC in 2025, with full roll-out expected by 2027. As with the NSC, the Concept Document does not explicitly refer to the GEC, although i t is referenced in supporting material. Including the GEC in the National Student Assessme nt Framework would signal its 7

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

importance and clarify its key features, including the different assessment components and how they contribute to students’ final marks. The OECD has conducted a review of the GEC and recommended, among other measures, that the Grade 9 curriculum tests focus on core subjects in order to prioritise teaching and learning time. It has also recommended reconsidering the current teacher assessment of students’ 21st-century skills and instead assessing these through the Integrated Project and student self-evaluation (OECD, forthcoming [8]). Policy recommendations to improve the design and implementation of South Africa’s National Student Assessment Framework Recommendation 1: Update the National Student Assessment Framework to include all major assessments and clarify their key characteristics, including their purpose and use of results South Africa’s draft National Student Assessment Framework has several strengths. The 2022 Concept Document signals a commitment to improving assessment practices, outlines how assessme nt data are expected to inform interventions at national, district and school levels, and aims to align assessment more closely with national education objectives. The PowerPoint presentation that the De partment of Basic Education subsequently developed reflects its evolving vision for the Nation al Student Assessment Framework that includes all assessments from Grades 1 to 12. Together these documents have informed policy dialogue with education stakeholders. However, the Concept Document provides limited clarity on how data will be used to monitor progress towards the objectives set out in the Department of Basic Education Strategic P lan 2025–30. There are also notable gaps. The document does not specify the frequency of the SASE, nor does it refer to the new

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

School-based assessments School-based assessment (SBA) – assessments developed and administered by teachers at the school level – plays a central role in South Africa’s assessment system, particularly in th e absence of full-cohort standardised student assessments. SBA contributes to students’ end -of-year grades across all school phases, though its weighting decreases as students progress. In the Foundation Phase ( Grades R –3), SBA accounts for 100% of grades. In the Intermediate Phase (Grades 4 –6), SBA contributes 80%, with 20% from standardised assessments. In the Senior Phase (Grades 7 –9), SBA contributes 60%, while standardised assessments account for 40%. However, evidence indicates considerable variation between SBA and national standards. This variability poses a particular challenge, as endof-year grades carry significant stakes for students, determining their promotion to the next grade. While the Concept Document and supporting material recognise the importance of both formative and summative SBAs, they provide limited information on their key characteristics, including the different forms of assessment (e.g., classroom tests and quizzes, written assignments, projects), the ir timing, and their purposes. Although the Concept Document refers to the Curriculum and Assessme nt Policy Statements (CAPS) for further details, the National Student Assessment Framework would benefit from greater clarity on the different forms of formative and summative SBAs, as well as from promoting a more balan ced approach between them. While the CAPS supports formative assessment, implementation in practice often leans more strongly towards summative purposes, particularly because SBA results contribute to term and final end-of-year marks. Teachers in South Africa devote a lot of time to marking and correcting student’s work; more than any other TALIS-participating country (OECD, 2025 [8]). Diagnostic assessments The Concept Document outlines plans to develop diagnostic assessments starting from 2022. Since then, the Department of Basic Education has developed diagnostic assessments for Grades 3, 6, and 9 to

However, the Concept Document provides limited clarity on how data will be used to monitor progress towards the objectives set out in the Department of Basic Education Strategic P lan 2025–30. There are also notable gaps. The document does not specify the frequency of the SASE, nor does it refer to the new qualification examination, the G EC or the N SC. While the Concept Document and supporting material recognise the importance of formative and summative SBAs, they provide limited infor mation on their key characteristics and instead refer to the Curriculum and Assessment Policy Statements for further details . Furthermore, in practice, classroom-based assessment in many schools remains strongly associated with summative purposes, particularly because SBA results contribute to term and year-end marks. What can South Africa do to further develop the National Student Assessment Framework? As the Department of Basic Education finalises the National Student Assessment F ramework, it will be important to include all major assessments and clarify their key characteristics, inc luding their primary purpose and the intended use of the data. The inclusion of the G EC and the N SC would signal their importance as key milestones in students’ progression. The National Student Assessment Framework can also serve as an important tool to promote South Africa’s SBA policy and reinforce the intent to support continuous, formative assessment. However, achieving a more balanced approach between formative and summative assessment in practice may require reconsideration of the policy, incl uding how SBA results contribute to term and year-end marks. Reducing the amount of formal or reported SBA could help free up teacher time which could instead be used to students who are struggling (OECD, forthcoming [8]). In addition, in order t o strengthen alignment with national education objectives, the National Stude nt Assessment Framework should explain how assessments will be used to monitor the key indicators and targets from the Strategic Plan 2025 –30, such as Outcome Indicator 3.1, which refers to average scores in the SASE for Grades 3, 6 and 9 (Department of Basic Education, 2025[1]). Explicitly mentioning these in the National Student Assessment Framework would promote greater policy coherence and enhance the

targets from the Strategic Plan 2025 –30, such as Outcome Indicator 3.1, which refers to average scores in the SASE for Grades 3, 6 and 9 (Department of Basic Education, 2025[1]). Explicitly mentioning these in the National Student Assessment Framework would promote greater policy coherence and enhance the use of assessment data in systemlevel monitoring, policy making and planning. Australia’s National Student Assessment Framework provides an example of how nationally agreed perf ormance measures can be systematically linked to assessment and reporting cycles (see Box 2). 8

DEVELOPING SOUTH AFRICA’S NATIONAL STUDENT ASSESSMENT FRAMEWORK © OECD 2026

Box 2. The Measurement Framework for Schooling in Australia 2025 – an example of linking assessments to national objectives The Measurement Framework for Schooling in Australia 2025 is a nationally agreed framework that defines how the performance of Australia’s school system is measured and reported to the public. It sets out key performance measures against which progress towards national education goals — as articulated in the Alice Springs (Mparntwe) Education Declaration — is monitored. The table below provides an example how performance measures are linked to specific assessments and reporting cycles. Measures Source(s) Frequency 2023 2024 2025 2026 2027

Student achievement: NAP – Literacy NAPLAN mean scale scor