Pupils’ scores show change is possible
The Human Sciences Research Council released the results of South Africa’s participation in the Trends in International Mathematics and Science Study (Timss) 2015 on November 29 2016.
Timss was first conducted in South Africa in 1995, and thereafter in 1999, 2003, 2011 and 2015. The most recent study provides an opportunity for South Africa to assess its achievements and to examine how the home and school contexts influence educational achievement.
South Africa participated at the grade nine level in 2003, 2011 and 2015.
The trend data provides us with a unique opportunity to plot the educational achievements during this period. The changes in school mathematics performance provide a measure of whether the historical differences in the system are shifting in the right direction.
Thirty-nine countries participated in Timss 2015 at the grade eight or nine level. As a low-performing country, South Africa tested at the grade nine level.
The top five ranked countries for mathematics in 2015, as in the previous studies, were from East Asia — Singapore (scoring 621), Republic of Korea (606), Chinese Taipei (599), Hong Kong (594) and Japan (586). The five lowest-performing countries were Botswana (391), Jordan (386), Morocco (384), South Africa (372) and Saudi Arabia (368) — countries from Africa and the Middle East.
Although South Africa continues to perform at the lower end of the rank order of countries, the more interesting story is the improvement in mathematics scores since 2003. The national score for mathematics remained the same for Timss 1995, 1999 and 2003.
By contrast, the 2003 mathematics score of 285 improved by 67 points to 352 in 2011.
This trend has continued to 2015, and the mathematics score increased by a further 20 points to 372.
This means educational achievement improved by 87 points for mathematics from 2003 to 2015, and this change translates to an improvement of just over two grade levels.
Of the 25 countries that participated in Timss 2003 and 2015, 19 improved their mathematics scores and five countries’ performance declined.
South Africa showed the biggest positive change of 87 points. Botswana improved by 22 points for mathematics, England improved by 20 points and the scores of Egypt and Jordan decreased. Over the 20-year period, from 1995 to 2015, the South African educational system improved from a “very low” level in 1995, 1999 and 2003 to a “low” level in 2011 and 2015.
But there is still a way to go to attain the desired achievement levels. The country should aspire for an improvement of at least 35 points in the Timss 2019 cycle and thus pass the 400-point mark.
Although overall achievement is low, there is a pocket of very high mathematics performers. One percent of South African pupils (mostly from independent and fee-paying schools) scored in what Timss defines as the advanced category (a score higher than 625). This group is part of the global 5% who achieve at this level. None of the other lower-performing countries achieved at this level.
The Timss data allows us to estimate provincial performance. Gauteng and the Western Cape are the top performers, followed by Mpumulanga, KwaZulu-Natal, the Free State and Northern Cape, which achieved similar scores. The three lowest-performing provinces are Limpopo, North West and the Eastern Cape.
All provinces, except the Western Cape, increased their achievement scores from 2003 to 2015. The provinces that showed the greatest improvement in mathematics scores between these two cycles are Limpopo by 117 points, Gauteng by 105 points, Eastern Cape by 96 points and KwaZulu-Natal by 91 points.
The mathematics scores in the Western Cape decreased by 23 points from 2003 to 2015.
Over this period, there have been changes in provinces that go beyond school and classrooms that may have affected provincial scores.
The score improvement of especially the lower-performing provinces has decreased the differences between the highest- and lowest-performing provinces. In 2003, the average mathematics score difference between the highest- and lowest-performing provinces was 170 points; this difference decreased to 62 points in 2015.
The improvement by the lower-performing provinces points towards more equitable achievement across the provinces.
South African achievement continues to remain very unequal, with the mathematics achievement scores in no-fee (quintile one, two and three) schools at 341, in fee-paying (quintile four and five) schools at 423 and in independent schools at 477.
About 80% of pupils at independent schools, 60% of pupils at public fee-paying and 20% of pupils at public no-fee schools achieved mathematics scores above the minimum level of competency of 400 points.
South Africa has high levels of poverty, inequality and unemployment. These social characteristics have an effect on education quality and there are high levels of variation between schools.
Although high-income countries focus on interventions inside classrooms to improve subject matter knowledge and achievement scores, low-income countries have to respond to two challenges: on the one hand they must focus on what happens inside classrooms and schools to improve teacher and pupil mathematical knowledge. On the other hand they must identify the effects of the learning and teaching contexts on educational achievement.
There are many home environment factors that influence achievement. In 2015, four out of 10 pupils came from households in which a parent or caregiver had achieved education above grade 12 (the statistic in no-fee schools is 33% and 63% in independent schools).
Pupils in those households with a higher education level score 43 points higher in mathematics than those households with grade 12 or less education.
Nearly 95% of pupils from fee-paying and independent schools have access to tap water and flush toilets whereas only 64% of pupils in no-fee schools have access to tap water and 44% have access to flush toilets.
Pupils in households with flush toilets score 56 points higher in mathematics, and those with tap water score 54 points higher in mathematics than those who do not have access to this infrastructure.
One-third of all pupils indicated that they frequently spoke the language of the test at home (English or Afrikaans). Pupils who frequently spoke the language of the test at home scored 60 points higher than those who spoke it less frequently.
These correlations give us an indication of factors associated with educational achievement and further regression analysis is needed to measure the strength of these relationships.
School conditions also affect achievement. School violence continued to be a concern and 17% of pupils reported being bullied at least once a week. This is double the international average of 8%. Pupils who do not experience bullying score 68 points higher than those who do.
Resources are important for teaching and learning: 82% of mathematics pupils and 69% of science pupils reported having their own textbooks.
There is a positive association between pupil achievement and the emphasis placed by schools on academic success. Pupils attending these scored 34 points higher in mathematics.
There is also a relationship between pupil absences and achievement, with two-thirds of grade nine pupils reporting that they are hardly ever absent. This group scored 60 points more than those who are absent often.
We are unable to report on teachers as the sample size of teachers is too small to generalise.
The increase of achievement scores over the 2003 to 2015 period shows that educational change is possible. The challenge for South Africa is to maintain the upward trajectory in achievement and to accelerate the pace of change.
The urgency is greater in no-fee schools, who have shown gains since 2003 but are still at a very low performance level. The education system is now in a less fragile state and greater demands can be made to improve the school climate so that it is conducive to both teaching and learning.
The interventions to improve educational achievement must focus on what happens inside classrooms and how teachers, with the requisite content knowledge, engage pupils to build their mathematics knowledge.
There isn’t a single silver-bullet solution. The provinces should identify the different groups and devise strategies to support both the low- and high-performing schools. Our analysis reveals that the more “good” conditions there are in the home and school, the higher the achievement scores.
The authors are researchers for the Trends in International Mathematics and Science Study at the Human Sciences Research Council.