South Africa: Research Overview, Education Policy Unit, University of the Witwatersrand

CREATE research in South Africa involved mixed methods and two rounds of data collection. Case studies were carried out in eight schools in the Ekurhuleni South district of Gauteng and six schools in the Dutywa district of the Eastern Cape, with a focus on learners in Grades 1, 3, 5 and 7 in 2007, and Grades 2, 4, 6 and 8 in 2008. Key baseline data was collected, including copies of registers, repetition data, new admissions, academic records, the Annual Schools Survey and school policy documents.

Interviews were carried out with school principals, mathematics and English educators and district officials. A baseline secondary analysis provided district-level indicators on learners’ degrees of vulnerability, with regard to over-agedness, repetition and drop-out. 1,121 Learner Profile Cards (LPCs) were fully completed by Ekurhuleni South learners and 596 by Dutywa learners. In addition, 61 of the approximately 150 questionnaires administered to parents of Ekurhuleni South learners deemed vulnerable by teachers could be matched to the LPCs, and 87 questionnaires were completed by parents of potentially vulnerable Dutywa learners. More information was gleaned through ‘Day in the Life’ activities, which involved shadowing one class in each grade through an entire day, and school and classroom observations were also undertaken. A sample of children were tested to establish their levels of achievement. Community fieldworkers were hired to search for out-of-school children between the ages of 7 and 15 as well as older youth who had not completed Grade 9.

Some Findings

Poverty is closely associated with educational exclusion in South Africa. One indicator of the extent of poverty is the fact that, in 2007, 64.5% of children aged 0-6 years received a child support grant, with 1.6% also receiving a care dependence grant and 0.2% a foster care grant. If receipt of the grant is recognised as an indicator of poverty then most children are in poor households. This of itself does not result in exclusion from the basic education phase since the majority of learners are enrolled up until the end of Grade 9. Nevertheless, poverty is associated with hunger, and hunger affects school attendance and academic performance. In 2003 children in 24% of households were always, often or sometimes hungry (DoE, 2006b:21). Poverty also makes the indirect costs of education, including uniforms, books, stationery, examinations, transport and the opportunity costs of education for older children more difficult to bear (Motala et al, 2007; Sayed and Motala, 2009).

Schools themselves play a big role in encouraging or discouraging access. The persistence of racism, sexism, bullying and xenophobia does not make some schools very inviting places, and this is compounded where there is poor quality of teaching and learning. The poor state of school infrastructure, shortage of classrooms, lack of decent toilets and play-grounds detracts from a healthy learning environment. The absence of state services to support schools in terms of social and psychological issues places increased burdens on teachers who are already overloaded (Williams, 2010; Motala and Dieltens, 2010).

Parents and guardians are not always able to provide the necessary background and knowledge of schooling to support their children, and many households are fractured. More educated parents are likely to encourage learning and to send their children to higher performing schools (Motala and Dieltens, 2010).

With unemployment hovering around 25% (2009 figures), there appear to be few economic rewards for remaining in school, let alone completing Grade 9, unless access to higher education is probable. Prior to the introduction of no fee schools, the second most important reason (after fees) given by learners as to why they remain out of school is that it is useless or uninteresting: a survey in 2004 revealed that almost 10% of learners overall, and more boys than girls (13.5% as against 6.5%), hold this view of the value of education (OECD, 2007:50). Our case study data confirmed that many learners had doubts about the value of remaining at school if they were unlikely to pass matriculation.

Over-agedness is a problem throughout the system, but especially in the higher grades. Over-age entry to schooling is being addressed by the age-grade norm policy. An unintended consequence is that many learners who would have repeated at lower grades are being allowed to progress, and some repetition is being deferred to higher levels of schooling (Motala, Dieltiens and Sayed, 2009; Taylor et al., 2010). The enrolment of under-age learners into Grade 1 despite policy continues to be a feature and is used as pre-school provision (Taylor et al., 2010). Age appropriateness has a distinct gender bias with more female learners being age appropriate than boys by Grade 9 (Motala, Dieltiens and Sayed, 2009).

The direct costs of education are being addressed through ‘no fee’ schools, but the indirect costs – of transport and uniforms in particular – are still a huge burden for poor households. Moreover, the persistence of fee-charging government schools alongside ‘no fee’ schools helps to sustain a class-differentiated two tier education system (Motala, 2008). Private schools make little contribution to overall enrolments (less than 5%) and most are high cost.

Levels of achievement are very low, amongst the lowest in the region. Children performed very poorly on CREATE attainment tests. Learners performed way below their expected levels in the tests, over-age and under-age learners were worse off in terms of their performance, prior learning for the majority of learners was poor i.e. they were not on the expected level for the grade (Gilmour et al., 2009; Pereira, 2010; Taylor et al., 2010)

Distance from school as well as schools’ official language of learning and teaching limit school choice, forcing some learners to travel long distances to other schools. Inadequate mastery of the language of learning and teaching is also a major factor in the abysmally low levels of learner achievement; yet many parents prefer (with their children’s concurrence) for their children to be taught in the second language of English by teachers who are themselves second-language speakers of English (Motala and Dieltens, 2010; Lafon, 2009; Alexander, 2010).

CREATE research found very little actual teaching and learning taking place in case study schools. Lessons often started late, much time is spent maintaining order, teachers do most of the talking and learners are passive and contribute little. The absence of writing and written work in classrooms was striking, rote learning and chorusing of lessons was common and coverage of the curriculum was very uneven (Letatsi, forthcoming). Corporal punishment is common and continues to be a feature of many schools and at different levels despite being formally proscribed.

Figures from the Community Survey in 2007 indicate that they are about 386,000 children who are out of school (Shindler, 2010). The great majority of these learners have dropped out. Much smaller numbers are children who have never been to school. CREATE research highlights specific factors that are correlated with exclusion including disability, household structure, poverty and lack of access to social grants. All of these increase the vulnerability of children to exclusion.

Some Policy Messages

  • Schools in townships and rural areas need to be made more welcoming in terms of infrastructure and facilities, pedagogy, care and community service. Aside from encouraging more active civic participation in schools and ensuring that teachers and principals always treat parents with respect, more attention could be given to planting trees, building playgrounds, painting murals, fixing desks and chairs, involving Community Development Workers in after-school activities, or paying stipends to unemployed matriculants to read to learners after school and to coach sport.
  • Language policy provides for 11 languages to be used but in practice English tends to be chosen most often as the language of learning and teaching (LOLT), along with Afrikaans, despite rhetoric of equality regarding the other official languages. School language policies impose limits on school choice and need clarifying and making more consistent with the government’s broader Language-in-Education Policy. Much learning takes place using languages that are not the LOLT, but this is often unplanned and not reflected in learning materials.
  • The continued use of corporal punishment often appears counter productive since it can act as a disincentive to attend school and may be used inconsistently. Though it is proscribed it continues to be widely used. Steps should be taken to end this practice.
  • The quality of teaching needs to be improved, through training as well as financial, practical and moral support for teachers. Classroom practice, pedagogical knowledge and in-service training need to take account of the gaps in teacher’s knowledge and plan for this.
  • Teacher accountability continues to be key. The rights of teachers to strike in support of better wages and working conditions is enshrined in law and in the constitution. Historical precedents have meant that school inspection can be infrequent. The low levels of numeracy and literacy scores in the basic education phase evidenced in both local and international benchmarking and the poor matriculation results (school leaving exam rate – 62% in 2009) indicates that teachers at the school level need to be made accountable for what they are delivering and what the outcomes are. CREATE research has illustrated that on too many occasions teacher contact time was limited, teachers were not present in the school during the school day or teachers were present at the school but in staff rooms or basking in the sun instead of teaching.
  • Repetition as remediation must be used as part of a clear pedagogic strategy; formative assessment should be promoted to give diagnostic insight into learning needs; on schedule progression accompanied by achievement consistent with age grade norms should be the expectation for all learners (Taylor et al., 2010).
  • The reality in South African classrooms is that there is often a wide range of ability within the same grade and that monograde teaching is the normal practice despite the fact that there are small schools with too few teachers and classrooms for every child to be taught in a reasonably sized grade group. Multigrade teaching and learning should be made more widely available.
  • There is a need to track learner’s migration, since it is common for children to travel long distances to school and to change schools. Migration can take many forms that include rural-urban, inter-urban and inter-school movement, as well as migration between provinces and from outside the country. Inter-school migration is likely to be affected by affordability, distance, access to transport and the perceived quality of education, as well as school language policies, ethos and reputation. It can contribute to age in grade slippage depending on how transitions are managed.
  • Information at all levels is lacking that would allow accurate tracking of first registration, progression, repetition and drop-out, and of achievement. Child identity numbers have been introduced (e.g. in Western Cape) along with regular assessment using standardised instruments. This practice should be extended to all provinces.