Building better: Why tracking quality is part of strengthening early childhood development in SA

4 BUILDING BETTER: WHY TRACKING QUALITY IS PART OF STRENGTHENING EARLY CHILDHOOD DEVELOPMENT IN SOUTH AFRICA

The earlier we invest in children’s development, the better. In South Africa, this means improving both access to and quality of learning programmes: about a third of the country’s 3–5-year-olds do not attend any form of an early learning programme (ELP) and of the children who are enrolled, about 60% are not developmentally on track.

Programme quality has two dimensions. The first, structural quality, relates to the physical setting in which children learn. The second, process quality, refers to how a programme creates opportunities for children to interact with the world around them. Research shows that this has a greater influence on children’s development than structural quality alone.

To improve process quality, it needs to be measurable in the first place. However, objectively measuring an abstract concept such as quality is tricky — even more so when the aim is to link findings on a programme’s offering to children’s developmental progress.

This learning brief examines how the process of developing an ELP quality assurance and support system unfolded, the progress achieved and the challenges ahead.

THE RIGHT INVESTMENT FOR REAL REWARDS

Investing in young children’s development is smart, research shows.1 Offering children structured and stimulating early learning opportunities capitalises on their natural responsiveness to the world around them — how they explore it, their role in it and how they engage with situations and people. This helps set them up to become productive citizens who contribute to the economy, create nurturing environments for their children and improve human development trajectories. The earlier the investment, the better the return.2

Yet in many developing countries, the environment to foster early childhood development (ECD) is not as strong as it should be. Although around 62% of the world’s children are enrolled in ELPs, children from low-income countries make up only a third of that group, with access even in that subset varying based on socio-economic status or geographic location.3 Similar trends are seen in South Africa: about a third of the country’s 3–5-year-olds do not attend any form of ELP, the 2021 ECD Census showed.4 In addition, of the children who are enrolled, about 60% are not developmentally on track, the 2024 Thrive by Five Index found.5

Strengthening ECD requires not only that more children have access to ELPs, but also that they’re exposed to quality learning. This means programme quality needs to be measurable in the first place – because only then is it possible to identify shortcomings, develop support and gauge whether interventions were successful in improving outcomes.

Research shows that the quality of an ELP has two dimensions.6 One relates to assessing structural aspects such as the physical state of a facility, group sizes, staff-to-children ratios and health and safety standards. These factors all help create conditions that allow children to learn well — and are fairly easy to quantify and compare against predefined benchmarks.

The other dimension, process quality, is harder to pin down. It focuses on how a programme helps children learn and develop through creating opportunities for interaction with other children and adults, their physical surroundings and materials, as well as their families and communities. Research shows that this category has a greater influence on child development.7

Objectively measuring quality can be tricky. How can a score be assigned to something that is descriptive in nature? And how can such scores be used to support and improve early learning,

1 Heckman, J. J. 2008. Schools, skills and synapses. Economic Inquiry 46(3), p. 289. Available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC2812935/ Bendini, M. M., et al. 2022. From evidence to effective policies: How to invest in early childhood education to nurture children’s potential - Overview. Washington, DC.: World Bank Group. Available at: http://documents.worldbank.org/curated/ en/099925111162229966

2 Heckman. Economic Inquiry, p. 289. Available at: https://pmc.ncbi.nlm.nih.gov/ articles/PMC2812935/

3 Bendini, M. M., et al. How to invest in early childhood education to nurture children’s potential – Overview. Available at: http://documents.worldbank.org/curated/ en/099925111162229966

4 Department of Basic Education. 2022. ECD Census 2021: Report. Pretoria: DBE. Available at: https://datadrive2030.co.za/wp-content/uploads/2022/09/ecdc-2021report.pdf

5 Giese, S. et al. 2025. Thrive by Five Index 2024: National findings. Pretoria and Cape Town: DBE and DataDrive2030.

Maldonado-Carreño, C., et al. 2022. Measuring the quality of early childhood education: Associations with children’s development from a national study with the IMCEIC tool in Colombia. Child Development 93(1), pp. 254–268.

THE MAKINGS OF A QUALITY ASSURANCE AND SUPPORT SYSTEM

The basis for a national QASS for early learning was set out in the National Integrated ECD Policy published in 2015.8 It called for a tool that could support programmes’ ongoing self-assessment, direct them to support and encourage the attainment of external accreditation.

Although government is mandated to support efforts to improve the quality of care and learning, capacity constraints in the public sector can slow progress. In addition, as the ECD sector is largely run by private providers, the wisdom and experience of social partners can’t be ignored. These realities provide an opportunity for collaboration between government departments and non-state actors such as NGOs, civil society and private funders.

Clearly defining the responsibilities of different role players has helped capitalise on their respective strengths. For example, as the mandated owner of a national QASS, the Department of Basic Education (DBE)9 is positioned to develop an enabling policy environment for quality assurance and support, establish a framework within which the system can operate, and coordinate the operations and resources needed for implementation, particularly through provincial departments. With this in place, the expertise of NGOs and other private sector actors can be leveraged to handle the operational development and rolling out the system, along with monitoring how it’s being used to facilitate further refinement.

Ilifa has worked with the DBE to bring together contributors from different areas of expertise, managing the various steps of design, development and evaluation, and facilitating dialogue between multiple stakeholders.

“The government’s responsibility for introducing and implementing a quality improvement system stems from the ECD policy, which states that regulated ECD standards for registration are important but insufficient,” explains Shakira Maharaj, project manager for quality and human resources at Ilifa.

Maharaj says the bigger picture of tracking quality is to help children reach their full potential: too many are not meeting the developmental milestones they should be.10 “We’re not going to see children thrive if we just support the basics around health and safety and programme delivery. Research has told us that one of the most important components for improving child outcomes is how the practitioner engages with children, and a lot of that is determined by quality standards. That’s not an environmental health and safety thing; that’s about how programme delivery is happening,” she says.

To that end, the sector must develop a shared understanding of quality, Maharaj explains. As a first step in designing the QASS, it was therefore necessary to define the quality standards, how to measure them and how to respond to them to better support principals and practitioners so that children can thrive.

Maharaj says developing such an assessment system has been a long journey, and “the DBE has been incredible in driving this process as a priority” by welcoming input from experts in various fields: statistics, psychometrics, ECD, data analysis and more. The shared thinking eventually culminated in an initial set of tangible, measurable standards in 2021, organised across five domains to provide a well-rounded picture of how a programme operates:

Learning programme looks at how learning is facilitated in the classroom and the materials or resources available to help children play and learn.

Staffing, management and leadership focuses on practitioner competencies and experience and opportunities for further development and training.

Parent and community engagement considers how programmes reflect what caregivers teach children at home.

Inclusiveness looks at how classrooms support everyone’s participation, regardless of ethnicity, ability, language, gender or socio-economic circumstances.

Nutrition, health and safety describes how programmes promote an environment that fosters safe and healthy development.

8 Department of Social Development. 2015. National integrated early childhood development policy. Pretoria: DSD.

9 Government oversight for ECD was transferred from the DSD to the DBE in April 2022.

To arrive at an objective measurement of the quality of learning offered by an ELP, each statement within a domain is scored as a 1, 2 or 3, representing three levels of quality: basic, emerging and enhanced. Scoring all the statements across each of the domains can then give a general picture of a programme’s strengths and areas for improvement. This allows the quality assessment to be used as a developmental instrument rather than a scorecard that penalises ELPs for non-compliance. Maharaj says this approach is one of the core principles of South Africa’s emerging QASS: “We want the

tool to meet programmes and practitioners where they’re at and help us figure out how we can take them on a journey of support.”

That’s why a package of support was developed alongside the assessment tool, describing the roles and responsibilities of different actors, as well as the resources, materials and supportive processes needed to improve the quality of early learning programming.

A DECADE OF DEVELOPMENT THINKING

National Integrated ECD Policy released

Early research about tracking ECD quality

Draft version of QASS put forward, together with a package of support

2023

Talks around an Early Childhood Care and Education (ECCE) Outcomes Fund set up by the DBE and donor organisations, and the need for a quality assessment system to track impact

Validation study

First roll-out as part of the ECCE Outcomes Fund

TESTING THE TOOL

To see how the proposed assessment tool worked in practice, it was field-tested on a small scale with about 20 sites in 2023. Once the results were in, the system would be refined and then trialled more extensively — or at least that was the plan.

But then talks between the DBE and donors began about launching an Early Childhood Care and Education (ECCE) Outcomes Fund,11 a nearly R500 million collaboration between the DBE, the Education Outcomes Fund (EOF) and a coalition of local and international donors and implementation partners. It is set to run over three years and expand access to quality early learning for more than 115 000 children in the Eastern Cape, KwaZulu-Natal and Limpopo, provinces where access to quality early learning is low. Implementing partners will either build new ECD centres in the communities in which they’re tasked to work or help 2 000 existing facilities meet the requirements for attaining Silver registration status and so become eligible to access the government’s per-child per-day subsidy, as well as funds for infrastructure development (on page 24).12

The ECCE Outcomes Fund uses an outcomes-based funding model, meaning that payments to implementing partners are made only once outcomes linked to real-world change — such as improved literacy or school readiness — can be demonstrated.13 An objective, reliable and valid assessment tool is needed to verify that an implementing partner’s work has delivered the intended results. This is exactly the purpose of the QASS. The development team therefore recognised an opportunity to run a validation study to test whether the system could be trusted to deliver the insights it was designed

to generate — particularly in understanding how the quality of a learning programme translates into children’s ageappropriate development and readiness for school.

With the help of a team of data analysts, statisticians and ECD experts, the standards included in the QASS were translated into a survey instrument, which could be tested in different ways to determine its reliability and validity.

11 Donors include The Lego Foundation, Yellowwoods, FirstRand, the Oppenheimer Memorial Trust and the Standard Bank Tutuwa Community Foundation. The five implementing partners are Ntataise, SmartStart, ELRU, Impande, and The Unlimited Child.

12 ECD Info Hub. 19 December 2025. Transforming early learning for 115 000 children with R469 million. Available at: https://ecdinfohub.org/thrive-by-five-report-key-findingsand-recommendations-copy/

Education Outcomes Fund. 2025. Harnessing the potential of outcomes-based financing to improve early childhood outcomes. Available at: educationoutcomesfund.org/post/harnessing-the-potential-of-outcomes-basedfinancing-to-improve-early-childhood-outcomes

For more about outcomes-based financing, see: Gustafsson-Wright, E., Gardiner, S. and Smith, K. 2016. Ensuring effective outcomebased financing in early childhood development: Recommendations to the International Commission on Financing Global Education Opportunity. Washington, DC: Brookings. Available at: https://www.brookings.edu/articles/ensuring-effectiveoutcome-based-financing-in-early-childhood-development/

Guerrero, A. 2025. Outcomes-based financing in the new financing for development architecture: Lessons and opportunities for governments, development partners, and multilateral organisations. Paris: OECD.

13 Education Outcomes Fund. No date. Work. https://www.educationoutcomesfund.org/work

Education Outcomes Fund. 2025. Harnessing the potential of outcomesbased financing to improve early childhood outcomes. 18 Dec. educationoutcomesfund.org/post/harnessing-the-potential-of-outcomes-basedfinancing-to-improve-early-childhood-outcomes

Did you know?

In statistics language, reliability means testing whether a research instrument yields the same results when administered by different people, while validity refers to whether the information gathered from the tool really answers the research questions.

For the validation process, 288 ELPs were assessed, with two field workers each running the survey at 234 sites to see whether the tool yielded reliable results. To check whether the tool yielded valid information about children’s outcomes, results were correlated with data from the Early Childhood Environment Rating Scale (ECERS) and the Early Learning Outcomes Measure (ELOM).14

The QASS ticked both these boxes, “something we were really excited about,” says Maharaj. “Seeing that the QASS could predict ELOM scores was a big win, because finding statistical correlations between quality assessment and child outcomes is generally very hard. In fact, when we told the analysis experts initially what we’re setting out to do, their advice was ‘don’t!’, because they said it’s likely that results will be quite vague and statistically hard to prove as definitive evidence of a relationship between quality and real outcomes.” But in the end, the results were positive.

NUMBERS THAT COUNT

5

Domains according to which ELP quality is assessed

FROM PAPER TO PRACTICE

The real challenge will be in implementing the system on the ground, says Maharaj. But the ECCE projects also present a valuable opportunity to gain insights into implementation and to learn from how NGOs respond to the assessment findings. “We’re excited to learn from implementing partners how programmes can be supported based on the assessment results, and then to work with the DBE to institutionalise the lessons into a public programme at scale.”

There are three important things to keep in mind when assessing how the QASS works in real life:

288

ELPs assessed as part of the validation study

1IMPROVING QUALITY IN ECD IS MULTIFACETED

Being able to assess an ELP is only one part of an ecosystem of ECD quality in South Africa; another important aspect is to help practitioners further develop their skills and competencies, and to support them by helping to create rewarding and nurturing work environments. “What we’ve seen in the process of developing the QASS, and also from other research, is that practitioners are the cornerstone of a learning programme and so, to move the needle on quality, we really have to support practitioners,” Maharaj explains.

Yet there’s substantial fragility in the sector, with nine out of 10 practitioners earning less than the minimum wage (R4 774 per month for full-time employment in 2024), the 2024 Thrive by Five Index shows. In fact, half of practitioners who took part in the Index survey indicated that low salaries were the biggest barrier to doing their job, while a quarter were planning to leave the ECD sector.15 In addition, only about a quarter of practitioners have a qualification beyond high school, and opportunities to pursue training and qualifications are often limited in the current ECD environment.

3

Provinces included in the validation study, namely KwaZulu-Natal, Mpumalanga and the Western Cape

The good news is that the quality standards as included in the QASS are referred to in the draft resource development plan being drafted by the DBE. Although this skills development plan is still a work in progress, it can help to concretise the relationship between developing practitioners’ capacity and programme requirements, says Maharaj.

QUALITY ASSESSMENT SHOULD DRIVE DEVELOPMENT, NOT BE A COMPLIANCE TOOL

The validation study showed that quality scores were distributed across domains and across ECD centres, showing the system is able to identify aspects that work well in a centre, as well as those that can be improved. That’s also good news, because the system appears to reflect local conditions well. “It’s a tool that’s been developed in South Africa for South Africa,” says Maharaj. “And that’s quite special, because many of the existing tools for ECD assessment speak to Global North conditions.”

Knowing that ECD centres operate at different levels of quality also raises challenges around how practitioners can best be supported. In theory, this could involve a process in which a centre engages with a quality coach after the initial quality assessment, followed by a period of support and a self-assessment before a formal re-evaluation to determine whether quality has improved. “But it’s going to be important to balance encouragement and recognition with identifying areas for improvement,” says Maharaj. “We really have to protect the QASS as a development tool.”

On-site coaching and mentoring have been shown to be effective, although the resource-intense nature of such interventions make them difficult to scale, explains Maharaj. It’s possible, though, especially when the expertise of the NGO sector can be leveraged, “because that’s really where the manpower and implementation experience sit.”

The role of NGOs was also highlighted in the recommendations that flowed from the Thrive by Five Index, noting that they “play a vital role in training and mentoring principals and practitioners, supporting curriculum

delivery and enabling access to quality learning materials. To strengthen this crucial support tier, greater and more sustained funding is needed.”16

Both Ntataise 17 and Impande 18 — implementing partners of the ECCE Outcomes Fund — are organisations with a proven track record in helping to upskill ECD practitioners. Ntataise, which has been working in this space since 1980 and has a network that reaches across six provinces, focuses specifically on helping women from marginalised communities acquire the knowledge and skills needed to establish and sustain ELPs in their communities.19 Impande similarly helps to empower women in rural areas who run small ELPs in their communities by, for example, hosting monthly learning groups for teachers to improve ECD programme implementation, helping practitioners at unfunded ECD centres receive an income through the Social Employment Fund, and collecting data about what centres need to improve the quality of their facilities.

MANY COMPLEMENTARY CHANGES ARE HAPPENING IN THE ECD ECOSYSTEM AT THE SAME TIME

After a sustained push over the last few years to improve the state of ECD in South Africa, many complementary systems are now being strengthened or introduced, and they all feed into each other. Take, for example, the DBE’s digital system known as eCares,20 which requires information regarding an ECD centre’s basic structural compliance as provided by the QASS. At the same time, the QASS can require information about centre registration as captured on eCares. “We have to be mindful that toggling a lever in one place can impact something else in the system, so there has to be a level of strategic thinking, as well as consideration for immediate operational aspects. That’s what’s really interesting and complicated and challenging about this work,” says Maharaj.

16 Thrive by Five Index 2024. Recommendations brief: Evidence to action. Pretoria and Cape Town: DBE and DataDrive.

17 Ntataise. No date. Ntataise. https://www.ntataise.org/

18 Impande. No date. Impande. https://impande.org/

19 Ntataise’s network has helped train more than 80 000 practitioners and reach ECD centres across Limpopo, Mpumalanga, Gauteng, KwaZulu-Natal, the Northern Cape and the Free State. Also see: https://www.ntataise.org/accredited-training-andqualifications/

20 The acronym eCares refers to the government’s Early Childhood Administration and Reporting System.

LESSONS LEARNT

It’s been a long road getting to a point where a workable version of the QASS can be used in a real-world setting at scale. Here’s what the process taught the team.

1

YOU CAN’T GET TOO ATTACHED TO YOUR ORIGINAL PLAN

It’s good to have a plan in place to guide how a process would work, but it’s also important not to hold onto it too tightly, says Maharaj. “When the ECCE Outcomes Fund opportunity came up, we had to let go of how we had initially thought of trialling the system. It was hard, because we had to rethink our timing, the budget, our capacity to manage the process well and the type of skills and thinking we needed to run the validation process. But when we look back at it now, it was the right decision to change course.”

2

DESIGNING SYSTEMS TO IMPROVE ECD QUALITY IS A COLLABORATIVE EFFORT

No single group has all the answers when it comes to developing something like a quality assessment system, which is why Ilifa brought people together and managed the process flow. “The QASS is a result of years of different thinkers supporting the DBE in their vision of strengthening ECD through a lot of compromise and robust debate. The process of getting different brains in the room to build what we eventually came up with was rewarding — even when difficult — and it’s something to take forward to other projects,” notes Maharaj.

NGOS ARE AN IMPORTANT PART OF THE ECD ECOSYSTEM

Even though the DBE is the mandated owner of the quality assessment process, NGOs are an important component of implementing the system on the ground and shaping the support response. “The ECCE Outcomes Fund will be an opportunity to test that relationship, at a large scale, especially because it will run in particular provinces [where access to quality early learning is lower],” says Maharaj. This could yield early insights into population-based planning for driving new programmes and improving existing ones, and how a public priority can be realised by leveraging NGOs.

4

DEVELOPING A QASS IS A RESOURCEINTENSIVE PROCESS

Getting to a point where a reliable and valid quality assessment tool is ready to be rolled out has taken years of work, lots of different expertise and considerable investment. The way the DBE has approached the development of the system, and seeing the end product as a shared tool, is exciting, says Maharaj. “It will ultimately be a public tool, and the hope is that organisations working in the ECD space will be able to use it to assess the impact of their support.”

WHAT’S NEXT?

The ECCE Outcomes Fund will act as a springboard for observing how the QASS works in practice at scale. Further work will also be required to see how the QASS can be integrated into the eCares system to ensure the QASS is supported by a strong digital backbone.

3 This is the learning experience of: