FROM LAGGARD TO WORLD CLASS

Research report no 13

Reforming maths andscience education inSouth Africa’s schools

This project has been funded by the Anglo American Chairman’s Fund, the

AngloGold Ashanti Fund, the BHP Billiton Development Trust, the De Beers

Education Trust, the FirstRand Foundation, the Joint Education Trust, the Liberty

Foundation, Murray & Roberts Holdings Limited, the Shuttleworth Foundation,

and the Zenex Foundation.

PREVIOUS TITLES

1. Post-apartheid population and income trends: a new analysis (September 1995)

2. South Africa's small towns: new strategies for growth and development (May 1996)

3. Cities and the global economy: new challenges for South Africa (October 1996)

4. Durban: South Africa's global competitor? (October 1996)

5. The East Rand: can South Africa's workshop be revived? (June 1997)

6. People on the move: lessons from international migration policies (June 1997)

7. People on the move: a new approach to cross-border migration in South Africa (June 1997)

8. Pretoria: from apartheid's model city to an African rising star? (July 1998)

9. South Africa's 'discarded people': survival, adaptation, and current policy challenges (October 1998)

10. Policy-making in a new democracy: South Africa’s challenge for the 21st century (August 1999)

11. Johannesburg, Africa’s world city: a challenge to action (October 2002)

12. Key to growth: supporting South Africa’s emerging entrepreneurs (June 2004)

Cover photograph by Gisèle Wulfsohn/South Photographs

Produced by Riaan de Villiers and Associates

CDE RESEARCHPOLICY IN THE MAKING

CDE RESEARCHPOLICY IN THE MAKING

13

FROM LAGGARD TO WORLD CLASS

Reforming maths and science education in South Africa’s schools

Abridged

Johannesburg November 2004

T H E C E N T R E F O R D E V E L O P M E N T A N D E N T E R P R I S E

Published in November 2004 by the Centre for Development and Enterprise

Pilrig Place, 5 Eton Road, Parktown, Johannesburg 2193, South Africa

P O Box 1936, Johannesburg 2000, South Africa

Tel 27-11-482-5140 • Fax 27-11-482-5089 • info@cde.org.za • www.cde.org.za

© Centre for Development and Enterprise

All rights reserved. The material in this publication may not be reproduced, stored or

transmitted without the permission of the copyright holder. The report may be quoted and

short extracts used, provided the source is fully acknowledged.

ISSN 1027-1406

CDE Research: Policy in the Making is a vehicle for disseminating research

findings and policy recommendations on crucial national challenges. Each

issue is based on in-depth research, including numerous specially commissioned

background research reports written by experts in the field.

SERIES EDITOR

Ann Bernstein

This report has been written by Ann Bernstein, Dr Tim Clynick, and Dr Robin Lee.

It is an abridged version of a full-length CDE research report entitled From laggard to

world class: reforming maths and science education in South Africa’s schools, written by

Dr Lee and Dr Clynick, with research assistance from Sean Willis. The full report is

available from CDE. Both documents have been edited by Riaan de Villiers.

Background research for this project was undertaken by Helen Perry,

Prof Gilbert Onwu, Prof Sarah Howie, Penny Vinjevold, Prof Charles Simkins,

Dr Robin Lee, Jocelyn Smith, Prof Diane Grayson, Aarnout Brombacher,

Dr Tony Morphet, Pam Watson, Dr Piet Human, Mike Erskine, Ruth Dube,

Makhosi Sigabi, Johnny Philander, Lucky Khumalo, Dudu Mkhize,

Dr Nopi Linkonyane, Dr Tholang Maqutu, Dr David Sirestarajah, and

Denrick Blaauw.

An advisory team helped CDE to manage this project. Its members were

Margie Keeton, Prof Sipho Seepe, Prof Charles Simkins, Dr Nick Taylor,

Penny Vinjevold, and Prof Diane Grayson.

In t roduc t ion 5

Quant i ta t ive research f ind ings 8

Qual i ta t ive research f ind ings 15

In ternat iona l exper ience 20

Poin ts o f depar ture 23

CDE’s recommendat ions 30

C O N T E N T S

Acronyms and abbreviations

CDE Centre for Development and EnterpriseFET Further education and trainingFETC Further education and training certificateGET General education and trainingGETC General education and training certificateHG Higher gradeNdoE National department of educationNGO Non-governmental organisationNSMSTE National Strategy for Mathematics, Science, and Technology Education NTF National Task Force on Maths and Science EducationOBE Outcomes-based educationSC Senior certificateSG Standard gradeTIMMS-R Third International Maths and Science Study

5

INTRODUCTION

The number of higher-grade Senior Certificate passes in maths and physical science in

South Africa was lower in 2002 than it was in 1991. Given the massive growth in the num-

ber of Senior Certificate candidates since then, it is clear that a national crisis has devel-

oped in maths and science education, with serious implications for economic growth. The

poor quality of schooling in these subjects is probably the single biggest obstacle to African

advancement in this country. – Main report

Education planners, educators, and parents in many countries are concerned about the

poor quality of mathematics and physical science education in their schools. Politi-

cians, economists, business people, and universities are concerned about the small num-

bers of learners leaving school with high enough grades to enrol for courses based on

maths and science at tertiary institutions. And everyone is concerned about the small

numbers of capable students being trained as educators in these two subjects.

South Africa is no exception. As long ago as the 1970s, key figures in the private sector

began funding maths and science projects in response to the disastrous impact of ‘Bantu

education’ on African learning and skills levels. After South Africa’s transition to democ-

racy in 1994, the education authorities launched numerous initiatives aimed at improv-

ing school-based maths and science education. In 1999, presidential priority was given to

these subjects. In June 2001, the national department of education (NDoE) drafted a

national strategy for maths, science and technology that resulted in the launch of the

Dinaledi (or ‘102 schools’) programme in September 2001. Provinces have implemented

their own initiatives, and the private sector and international funders have contributed

projects of their own.

Good work is being done, and initiatives are being successfully implemented. However,

the maths and science education system is still failing to deliver enough school-leavers

equipped with HG maths and science to meet the country’s needs. While in 1991 South

Africa produced 20 667 HG SC maths graduates, in 2003 it produced 23 412 – the first

time in over a decade that numbers exceeded the 1991 figure.

It is clear that a new approach is required if our schools are going to deliver in this vital

area. Whereas, over the past decade, the education authorities have generally tried to

change the whole maths and science education system at once, the private sector has con-

centrated on individual projects. Neither approach is achieving the desired goal of equip-

ping the school-based system of maths and science education to realise the full potential

of South African learners, and meet the country’s ambitions for the next decade.

A question of national concern

South Africa is significantly disadvantaged globally and in terms of its national priorities

by the poor performance of its maths and science education system. This is particularly so

in view of the fact that these subjects are increasingly important to any economy that

wishes to compete in the global economy. Competitive economic activity in the 21st cen-

tury will inevitably involve the extensive use of technologies that are ultimately based on

maths and various branches of science. South Africa cannot hope to develop these tech-

The maths and science

education system is

failing to deliver enough

school-leavers equipped

with HG maths and

science to meet the

country’s needs

Greater African

participation in the

economy is being held

back by the poor quality

of our schooling system

6

F R O M L A G G A R D T O W O R L D C L A S S

nologies – or even to productively apply technologies developed by others – without a large

and growing group of citizens with a sound maths and science education. And it cannot

hope to produce this core of specialists without effective school-based education in these

subjects by well-qualified, highly committed, and adequately remunerated educators.

Government policies in the areas of economic growth, black economic empowerment,

skills development, and other critical areas assume that the education reforms of the past

10 years have already dramatically increased the supply of black SC maths and science

graduates to higher education institutions and the public and private sectors – but this is

not the case.

Key government activities also require citizens capable of grasping the principles

underlying developmental policies and programmes. Citizens of a 21st-century democ-

racy cannot hope to contribute to or benefit from technology without a reasonable level of

maths and science education.

Private sector concerns in this area are equally fundamental. Most formal jobs in the

economy now require some competence in at least maths, and a higher level of mathe-

matical proficiency is, of course, essential in the finance, technology, and other sectors.

However, work in manufacturing, construction, retail, the services sector, and commer-

cial farming are increasingly dependent on maths literacy as well. Greater African partici-

pation in the economy is being held back by the poor quality of our schooling system.

The country’s top educational priority

International evidence, the opinion of domestic educational experts and practitioners,

and the views of parents and learners all confirm that maths and science are the compo-

nent of the education system that needs to be reformed most urgently. This view is rein-

forced by the fact that the private sector has committed substantial resources to maths

and physical science projects over the years, with considerable success at the level of indi-

vidual projects, but without having the critical mass to change maths and science per-

formance or participation. As yet, only a few large-scale ways to co-operate with govern-

ment in improving maths and science have been found and implemented successfully.

It is in the context of these realities that CDE has conducted a major privately funded

study of South Africa’s maths and science schooling system. After three years of intensive

research, CDE has produced a substantial report which, in many respects, represents the

most exhaustive analysis of this system yet produced. The study was prompted when some

of the country’s largest corporations expressed concern about the dwindling supply of

school-leavers with HG maths and science, particularly Africans, available to take up bur-

saries on offer for higher education.

The CDE study has been funded by ten donors: the Anglo American Chairman’s Fund,

the AngloGold Ashanti Fund, the BHP Billiton Development Trust, the De Beers Education

Trust, the FirstRand Foundation, the Joint Education Trust, the Liberty Foundation, Mur-

ray & Roberts Holdings Limited, the Shuttleworth Foundation, and the Zenex Foundation.

Research conducted for the project

Extensive research was conducted for this project. Twenty-seven background research

reports were written, and hundreds of interviews were conducted with national and

provincial education officials, academics and other analysts involved in education, repre-

We have continually

looked for what is

working, trying to

identify a sound

foundation on which

an improved system

could be built

7

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

sentatives of relevant NGOs, school principals, educators, learners, and others. The

research components include:

• a statistical analysis of performance trends in the SC examininations in maths and

physical science, 1991-2002;

• a more detailed analysis of the results of SC examinations in maths and physical science

in 1998, 2000, 2001, and 2002;

• compiling a national database of SC maths and science performance and participation

by individual schools across the nine provinces, and developing a set of criteria for

understanding and comparing the relative performance of all schools by various rele-

vant categories (home language, location, gender, etc);

• in-depth case studies of 13 schools in five different provinces which consistently per-

form well in SC maths and science;

• an examination of government and private sector initiatives in support of improved

maths and science learning and teaching;

• a review of the current system in the light of the competencies and morale of educa-

tors, curriculum reform, and the present SC examination; and

• a review of international experiences of system change, specialist schools, incentives

for educators, and how our performance in maths and science schooling compares

internationally.

The body of material produced in the course of our research is a valuable national

resource, which is now available to government and other interested parties. A number of

the analytical tools developed in the course of our research could play a valuble role in

helping policy-makers and planners to design new policy initiatives and programmes, and

monitor and assess their progress.

CDE’s approach

In undertaking this study, CDE has drawn on its years of experience in analysing and

assessing complex national policies and large programmes of delivery. We have been

guided throughout by the realities of reform and the difficulties of implementation in a

developing country recovering from the terrible legacy of apartheid. We have continually

looked for what is working, trying to identify a sound foundation on which an improved

system could be built. In the international literature, this is described as ‘seeking virtue’ in

an existing system.

It is a very different approach from one that says, everything is wrong with a given sys-

tem, and we must change it in its entirety. Experience over the past ten years in many sec-

tors of South African society has emphasised the importance of identifying ‘what works’

in this country, and incrementally building on that base in order to improve delivery in a

sustainable and more equitable manner. This philosophy of change has informed our

research as well as our final report (see box: In the midst of a national crisis, ‘virtues’ can be

identified, p xx).

In this abridged version of our full-length report, we will briefly review our key research

findings, both quantitative and qualitative; draw out key lessons from international expe-

rience; formulate points of departure towards a new approach; and put forward a set of

achievable, practical recommendations aimed at significantly increasing the numbers of

SC maths and science graduates within a limited time frame, which will, in turn, provide a

base off which to further expand access to quality education. If we act decisively, and with

If we act decisively, the

performance of the maths

and science education

system could be

dramatically improved

in a short period

8

F R O M L A G G A R D T O W O R L D C L A S S

the right mix of interventions, the performance of the maths and science education sys-

tem could be dramatically improved in a short period.

QUANTITATIVE RESEARCH FINDINGS

South Africa has a serious problem in the important maths and science compo-

nent of its education system.

In order to gain an accurate picture of the system’s performance, three factors need to be

considered: trends in overall enrolment, trends in pass rates, and trends in the absolute

numbers of passes produced. Ideally, trends should improve in all three areas, but espe-

cially in pass rates and numbers at the HG level. These school-leavers can enter higher edu-

cation, can start training as maths and science educators and, even if they are immedi-

ately employed, will be better equipped for on-the-job training. In other words, the

quantity of high-quality passes is the single most important trend. For the purposes of our

study, figures were compiled for a 12-year period, from 1991 to 2003.

Enrolment

Over the 12 years in question, total enrolment for SC mathematics increased from

135 659 to 258 323 (or 90,4 per cent). However, this growth took place at SG level only;

while SG enrolment increased from 82 028 to 225 033 (174,33 per cent), HG enrolment

plummeted from 53 631 to 35 959 (a decline of 32,95 per cent). HG mathematics is

essential for many higher education qualifications.

Similarly, while total enrolment for SC physical science increased from 84 019 to 151

791 (80,66 per cent), SG enrolment increased from 33 065 to 99 711 (174,34 per cent),

and HG enrolment only marginally from 50 954 to 52 080 (2,2 per cent).

CDE’s research has identified numerous positive elements, or

‘virtues’, in the existing maths and science education system

that can used as building blocks for a reform strategy. These

include:

• a national understanding that a problem exists;

• a commitment by government and private business to

allocate resources to addressing it;

• presidential interest in achieving delivery in this area;

• private and NGO sectors with a long history of concern

over and engagement with maths and science education;

• considerable potential in the current system to achieve

In the midst of a national crisis, ‘virtues’ can be identified

more HG passes;

• sound learning and teaching in some of South Africa’s

poorest schools and communities;

• a public recognition of excellent achievements under

adverse local circumstances;

• provincial energy, and innovative programmes;

• progress on a sound database of SC results, correlated

with other relevant variables; and

• an acceptance, signalled by the state-run Dinaledi

programme, that maths and science education needs

specialised attention. CDE 2004

One of the most worrying

aspects of maths and

science education is the

small proportion of

African learners who

pass these subjects

in the HG

9

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

Pass rates

The total pass rate in SC maths increased slightly from 47,9 per cent to 49,6 per cent.

While the SG pass rate declined from 48,7 per cent to 44,7 per cent, the HG pass rate

ncreased from 38,6 per cent to 66,1 per cent.

Similarly, the total pass rate in SC physical science increased slightly from 64,6 per cent

to 67,0 per cent. While the SG pass rate dropped from 67,1 per cent to 62,0 per cent, the

HG pass rate increased from 45,4 per cent to 50,1 per cent.

The increase in the HG maths pass rate, for example, seems impressive until one recalls

that enrolment in that subject dropped hugely over the same period. This shows that these

two sets of figures should be related to one another. When they are, they reveal a consis-

tent inverse relationship between enrolment and performance; the more learners enrol for

any of these three subjects, the worse their pass rate, and vice versa. Experts say this

demonstrates the widespread approach by school principals and educators pursuing good

SC results to discourage all but the most talented learners from enrolling for HG maths or

science. Thus the improved pass rates in HG maths and science are being achieved at the

expense of participation, and the improved participation in SG maths and science are not

being translated into results.

Number of passes

Perhaps the most direct measure of performance is the number of passes at SG and HG

level produced by the system every year. In respect of maths, the total number of passes

over the 12-year period has grown from 64 941 to 128 119 (or 97,28 per cent). However,

once again, the growth has only been in SG; while SG passes have increased from 39 028 to

99 426 (154,75 per cent), HG passes have only increased from 20 677 to 23 412 (13,22

per cent). In fact, in 2003 the system produced only 2 735 more HG maths passes than it

did in 1991 – only a fraction of what the country needs.

Similarly, while the total number of passes in phyical science has increased from

84 019 to 151 791 (or 80,66 per cent), SG passes have increased from 22 216 to 61 756

(177,97 per cent), and HG passes from 23 109 to 26 067 (12,80 per cent). Again, in 2003

the system produced only 2 958 more HG science passes than it did in 1991.

The pass rates in all four subjects have improved over the past three years. However,

experts attribute this largely to the introduction of the 25 per cent ‘continuous assess-

ment’ component of the SC, undertaken by schools themselves and not by independent

examiners.

African learners

One of the most worrying aspects of maths and science education is the small proportion

of African learners who pass these subjects in the HG. Our research shows that in 2002

only 4 637 African learners graduated in HG maths, amounting to 13,14 per cent of all SC

graduates, and 23,42 per cent of all HG maths graduates. Similarly, only 7 129 African

learners graduated in HG physical science – 14,06 per cent of all matric graduates, and

30,42 per cent of all HG science graduates.

In a comparison of the

performance of grade 9

learners in maths and

science in ten developing

countries, South Africa

scored lowest in

both subjects

10

F R O M L A G G A R D T O W O R L D C L A S S

Educators

Fewer SC graduates are enrolling for higher education courses leading to teaching qualifi-

cations in these subjects, so the entry of newly qualified maths and science educators is

not even keeping pace with retirements, retrenchments, and losses to other sectors, never

mind actually increasing the country’s resources in these subjects. In 2000, the number

of students at teacher training colleges was 56 per cent less than in 1994 (since then,

teacher training colleges have been amalgamated with universities). And, between 1996

and 2000, the number of education degrees awarded at universities and technicons

declined by 5,4 per cent.

The key issue is whether the country has ‘turned the corner’, and we can expect more

learners to achieve better marks in maths and science – particularly in the higher grade.

Most analysts do not expect this to be the case.

If we look at the system’s performance over the entire period from 1991 to 2003, what

can we say? The data as a whole show a sharp downward trend in the period to 2001, and

then a slight upturn, the causes of which are not entirely clear, and the duration of which

is still unpredictable. Nevertheless, over the entire period the increase in enrolment and

passes remains negligible for a system that is expanding overall, and a country that des-

perately needs major improvements. It is entirely unsatisfactory to be standing still over a

12-year period, whatever the reasons.

South Africa is in a worse position than virtually any other comparable develop-

ing country.

In order to assess South Africa’s performance in maths and science, this has to be com-

pared to those of other developing countries. In a comparison of grade 4 learners in 12

developing countries, conducted in 2000, South Africa scored lowest in numeracy, and

second lowest in literacy in English – a factor crucial to success in maths and science. (In

most of the other countries, As in South Africa, English is a second or third language, but

the most common language of learning and teaching.) In a second, related, study com-

paring the performance of grade 9 learners in maths and science in ten developing coun-

tries, South Africa again scorest lowest in both subjects.

South Africa’s challenge in maths and science has very specific origins related to the

country’s history of apartheid education. These are discussed in detail in the main report.

Yet the country is by no means alone in facing major challenges. CDE ’s international

research identified very few countries that are satisfied with their achievements in maths

and science education. Nevertheless, available comparative data indicates that South

Africa is worse off than virtually any other country. (Of course, there may be some coun-

tries that have not participated in comparative studies, and may have even less effective

education systems than ours.)

In 1998–9, the Third International Maths and Science Study (TIMMS-R) was conducted

in 38 countries. In a standardised test administered to grade 8 learners, South Africa per-

formed worst of all the other participating countries.

Using data drawn from the TIMMS-R study, CDE compared South Africa to ten similar

developing countries – Chile, Czech Republic, Indonesia, Korea, Malaysia, Morocco,

Philippines, Thailand, Tunisia and Turkey – which it thought would be a fairer compari-

son than the whole group of countries that included several developed nations. However,

Only 14 per cent of

schools in 2001 reported

that their maths and

science educators had the

minimum qualifications

required by the NdoE

11

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

even in this totally comparable sample, South Africa’s grade 8 learners were still the worst

performers, even though they were 1,1 years older on average than all other participants.

Most of these countries have similar social problems to ours which also impact on their

educational systems, such as a multiplicity of languages, massive income differentials,

sharp urban–rural divisions, and recent histories of conflict. Therefore, the factors that

have been used to explain our poor performance are not unique to this country, and other

countries with similar disabilities seem to be overcoming them more effectively than we

are.

The self-report elements of the TIMSS-R study give some possible reasons for our poor

performance. Our school year is substantially shorter than those of all the other countries

studied. We have the highest absentee rates for educators and learners. The time devoted

to instruction in maths and science is the least of all countries. Maths and science educa-

tors and learners spend significantly more of this time on repetition and homework than

their peers who spent more time clarifying key concepts or making steady progress

through a systematic learning curriculum or programme in the classroom. Effective utili-

sation of available resources is well below the average of comparable countries. And coun-

tries whose maths and science classes are as large as – and even larger than – our own

perform significantly better than we do.

In brief, South Africa faces a major challenge in bringing its maths and science educa-

tion system up to international standards.

Poor performance and low participation is a national issue, but this problem

manifests itself differently in every province and in every school.

CDE’s statistical analyses and development of SC maths and science results has made it pos-

sible to analyse participation and performance in these subjects so that we can identify the

problem nationally, provincially, and in individual schools. The research confirms that

these problems manifest themselves in different ways in every province and nearly every

school in the country.

Our research on school performance reveals several dimensions to this feature of the

maths and science schooling system. According to data in 2000:

• Almost one fifth of secondary schools did not offer SC maths and science at all – in fact,

the number of schools in this category appeared to be growing (by 7,3 per cent from

1998 to 2000)

• About one third of secondary schools offering maths and science at either SG or HG

achieved pass rates of 0–19 per cent.

• Only about half (2 929) of all secondary schools offered HG maths.

• Of these, 1 552 (or 54 per cent) achieved a pass rate of only 0-19 per cent. Total passes

achieved by these schools – 1 013 – constituted only 5 per cent of the total.

• At the other end of the spectrum, 701 (or 24 per cent) achieved pass rates of 80-100

per cent. Total passes achieved by these schools – 11 569 – constituted no less than 63

per cent of the total. This shows that there is a small core of good schools which consis-

tently perform at a very high level.

Very different trends emerge once the analysis goes beyond simple pass/fail grades, as pro-

vided in percentages by official data and debated in the media. Indeed, simple pass/fail

rates actually conceal a far more important trend, namely that the numbers of HG passes

have been declining for some time.

A large number of

learners with the

potential to succeed in

maths and science are

not getting the

opportunity to study

these subjects

12

F R O M L A G G A R D T O W O R L D C L A S S

Provincial variances are also significant; those provinces that had to integrate several

apartheid-era education departments (typically those provinces incorporating former

‘homeland’ territories) are most disadvantaged. The research has therefore underlined

that, even in these two subjects, we are dealing with a diverse and complex system, requir-

ing different policies and strategies in different provinces, different regions within

provinces, and in different schools.

Three key factors have emerged as major determinants of success in SC maths

and science.

Educator knowledge: The most important variable is the educational qualifications (content

knowledge) of educators, the pre- and in-service development of their teaching skills, and

their length of experience. Variations in these factors correlate significantly with varia-

tions in results, and indicate that any major intervention must first be directed towards

educators. This is an urgent priority, as only 14 per cent of schools in 2001 reported that

their maths and science educators had what government considers the minimum level of

qualification (SC plus 3,5 years of higher education). Given the shortage of suitable educa-

tors, fewer periods are allocated to these subjects in the school timetable, further com-

pounding the problem.

Language competence: The next variable is competence in one of the languages used for

instruction and examination purposes (mostly English). There is a significant statistical

correlation between marks achieved in the SC in the language of instruction and examina-

tion and achievement in maths and science. Empirical investigation in the 13 schools

studied by CDE confirms this correspondence, and the examiners, moderators, and mark-

ers who participated in the workshops on the SC examination agreed that this was the

case. The evidence is so strong that we will propose not simply maths and science reforms,

but maths, science, and language reforms.

School and classroom environment: The third significant variable is the nature of the school.

A specific set of school characteristics emerges very clearly as a precondition for success.

This finding has several dimensions. Statistically, the higher the number of candidates in

SC maths and science at a given school, the higher its success rate. Therefore, peer pressure

and teamwork obviously help. But the management of the school, standards of discipline

and orderliness (both physical and psychological), the commitment of educators, and par-

ticularly what happens in the classroom also come into play.

Many more learners could pass SC maths or science in the higher grade, but do

not enrol for either of these subjects.

Our research shows that a surprisingly large number of learners who could succeed at SC

maths and science do not enrol for these subjects at all, or enrol in the SG when their

marks indicate they could succeed in the HG. For example, our research shows that:

• In 1998, 41 per cent of learners who passed SG maths could have passed HG maths (this

calculation is based on an analysis of their aggregate SC mark). In 2000, the correspon-

ding figure was 56 per cent.

• In 1998, 50 584 SC candidates did not study maths at any level because they believed

or were advised that their aggregate mark was not high enough to support a pass in the

subject. In 2000 the corresponding figure was 56 195. However, their overall marks

CDE’s quantitative

research suggests that a

single, undifferentiated

policy to ‘improve maths

and physical science

education’ is

inappropriate

at this stage

13

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

suggest that they would have passed had they enrolled for these subjects. The numbers

are high: 53 283 more learners could have passed maths in 1998, and 49 439 in

2000. Some of these learners even had the potential to pass HG maths: 3 644 in 1998,

and 3 728 in 2000.

A similar if less dramatic pattern is revealed for physical science candidates. The implica-

tions are twofold: first, the country is committing major resources to huge numbers of

learners who, under current circumstances, have little if any chance of passing SC maths

and science, and should be guided on to alternative study paths. Simultaneously, a large

number of learners with the potential to succeed in maths and science are not getting the

opportunities to study these subjects which they deserve, and the country needs.

Whatever the reasons for this, South Africa cannot afford to continue to ignore this

wasted potential. This issue can be addressed immediately, as our recommendations indi-

cate.

There are too few excellent schools and too many bad schools in South Africa.

As we have shown our analysis of individual school performance reveals that:

• only half of all secondary schools offer HG maths;

• fewer than 20 per cent of schools with maths and science candidates attained pass

rates higher than 80 per cent in each subject;

• 24 per cent of schools offering HG maths achieved 63 per cent of the total number of

national passes.

• More than half of the schools offering HG maths – 1 552 – produce just 5 per cent of

the total number of passes.

• Between these two extremes are a further 676 schools with variable rates of success

ranging from a 20 per cent pass rate to close to 79 per cent. Together, they contribute

32 per cent of all HG maths passes.

What this means is that the system is highly skewed and highly vulnerable to any prob-

lems that may arise in a small number of schools. Resources need to be targeted carefully

and effectively. They must go to the top performing schools to ensure that their perform-

ance is maintained and that the largest possible number of learners can benefit from their

capacity to deliver. Then, schools near the top of the next category (pass rates of 60—79

per cent) can be identified and helped to move up to the 80 per cent plus category.

Attaining these results in schools arbitrarily ranked by performance remains a chal-

lenge, as all schools are different, and the causes of poor performance also differ. However,

ranking schools by performance relative to other schools will allow more specific sets of

appropriate interventions to be generated and implemented. In general, therefore, the

quantitative data generated by CDE makes it possible for resources to be directed strategi-

cally to various levels of achievement, so that excellence can be sustained and extended at

the same time that grave deficiencies are receiving attention.

There are many areas in which it is not necessary to act: reforms need not

change every part of the system at once.

CDE’s quantitative data helps to identify areas in which it is not necessary to act – for

instance, that of gender. A single national programme to increase female participation in

maths and science is not indicated. However, as in other respects, the data show that con-

There are many problems

in the public education

system, but there is no

general or fundamental

breakdown of our

schools or of maths and

science education

14

F R O M L A G G A R D T O W O R L D C L A S S

ditions vary in respect of gender participation as well, and that a programme aimed at

increasing the number of African female learners is urgently required in some provinces.

Similarly, under present conditions, access to a science laboratory is not correlated with

consistently high levels of success. This is not to say that this resource is not ultimately

desirable; however, new resources will be more immediately productive if they are directed

towards better educators and more textbooks.

Other kinds of interventions in specific provinces, regions, or schools might also be

needed, and CDE’s school performance index and other data now enable these to be identi-

fied down to the level of individual schools.

Conclusions from the quantitative research

CDE’s quantitative research suggests persuasively that a single, undifferentiated policy to

‘improve maths and physical science education’ is inappropriate at this stage. Any reform

programme must be based on the realities of maths and physical science performance and

participation patterns in individual schools. We require targeted, focused initiatives that

will build systematically on the existing ‘virtues’ in the schooling system.

There is a large pool of SC learners who have not enrolled for these subjects, but could

do so. It includes learners who are currently attending schools with non-existent or inade-

quate maths and science teaching capacity; it also includes those candidates enrolling at

SG when a significant percentage could pass at HG. Urgent steps are required to address

this wasted potential.

Many schools do not have the capacity and resources to teach maths and physical sci-

ence efffectively; however, a generalised support strategy for all schools is not called for.

Rather, the core of performing and now improving schools (many of which are located in

disadvantaged areas) that have always managed to generate significant numbers of HG

passes ought to be helped to maintain their remarkable performance.

Other schools that are on the brink of success due to hard work and effort should be

identified and helped to achieve specific goals. In both cases, building on ‘virtues’ identi-

fied in the system will achieve quicker and better results, and give much-needed momen-

tum to improving the performance of the education system as a whole.

Once we start to consider actual reforms, we will need to distinguish between maths

and physical science. Physical science is within the reach of many more candidates than

is maths. With better focus on the language of instruction and examination, improved

performance is more likely to follow for candidates in science as it is learnt through

description, supplemented by some observation. Better mathematics performance how-

ever must come from both improved conceptual and analytical skills: choices will there-

fore be required in terms of priorities which will have an impact on where resources are to

be applied.

We believe priority should be given to retaining effective and experienced maths, physi-

cal science, and English educators, while gradually expanding the supply by creating a

virtuous cycle of increasing numbers of maths and science HG passes, more graduates

entering higher education, and more suitably qualified maths and science educators. In-

service training programmes or professional development programmes should comple-

ment this strategy by improving current educators’ content knowledge.

CDE has serious concerns

about maths and science

education in grades 10

to 12 in the interregnum

before the introduction

of FET

15

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

QUALITATIVE RESEARCH FINDINGS

There are many problems in the public education system, but there is no general or funda-

mental breakdown of our schools or of maths and science education. This was confirmed

by our qualitative research, which also identified a number of ‘virtues’ in the public edu-

cation system. This is an obvious yet still surprising finding. We believe these virtues

should serve as the building blocks of any strategy aimed at improving maths and science

education. Whatever shortcomings exist can be remedied. We have derived this insight

from a number of important analyses.

Government initiatives

Our full-length report examines national and provincial government initiatives to improve

maths and science education since 1994, and we conclude that a concerned government

has invested significant resources, thought, and effort in this field. The level and quality of

this concern is an important virtue. However, these initiatives have not yet produced the

results commensurate with the time, energy, and resources invested in them. It is essential

to pinpoint the reasons for this. In brief, we believe these are: discontinuities of policy; too

many and overambitious policy changes; a lack of an overarching vision; and failures in

implementation (see box: Government initiatives, 1994–2003, p xx).

Some government initiatives, notably the National Strategy for Mathematics, Technol-

In general, government initiatives have been characterised

by:

• Grandiose policies: Attempts to change the whole

system, or large parts of it, all at once. All the

international research consulted by CDE shows that this is

a futile and ultimately counterproductive approach.

• Policy discontinuity: New policies conflict with,

diverge from, or abruptly discontinue initiatives already

launched, sometimes by the same department

• Undifferentiated strategies: Important subjects

such as maths and physical science are treated as if their

context and national importance are exactly comparable

to all other subjects; provinces are treated as if they are

all the same; and differentiated approaches to educators

and funding are positively discouraged.

• A lack of political will: Government has to establish

what its priority areas are in maths and physical science

education, and stick to its guns. For example, if planners

pander to local political considerations in selecting

Dinaledi schools, it will undermine this important

experiment in specialist maths and science schools.

Government initiatives, 1994-2003

• Inattention to maths and science educators as

a specific group: Real improvements in learning and

teaching have been slow in coming because of the absence

of a mechanism for testing educators’ content knowledge,

and of ongoing professional development. Also, despite

efforts to limit personnel costs, space must be created for

new entrants to the profession in these subjects, and ways

found to attract and retain new teachers.

• An inability to generate professional pride

and motivation: Reforms must pay attention to the

process of change; ie, educators should experience

change as important and something that makes a

difference to their lives as professionals; and should be

motivated by incentives to enter and remain in the

profession.

• Poorly planned, too rapid implementation:

Sufficient time must be allowed for maths and physical

science innovations to take effect; clear and achievable

outcomes must be identified; sufficient resources should

be allocated to achieving them; and progress should be

monitored. CDE 2004

Educators have not been

adequately trained in

how to administer

continuous assessment,

and the process is not

adequately monitored

and controlled

16

F R O M L A G G A R D T O W O R L D C L A S S

ogy and Science Education, have reported positive results. The Dinaledi (or ‘102 schools’)

programme is an important experiment which, though conceptually advanced, has not

been well implemented (see box: The Dinaledi programme, p 21). We are convinced that

improvements can be introduced without losing the gains that have already been made.

The new FET system

Our main report details the steps taken by the NdoE to establish a new school-based educa-

tion system. Significant design and implementation challenges remain in the area of the

new FET curriculum and the FETC that will finally replace the SC examination in 2008.

When the new FET syllabus is eventually implemented, beginning in grade 10 in 2006,

it will be a superior qualification to the present SC in many respects. The current HG and SG

maths will be replaced by two new subjects called mathematics and mathematical literacy,

and all FET candidates will have to study one of them. Mathematics is meant to be equal,

or even superior to, the current HG maths. However, if the current versions of the National

Curriculum Statements are taken as a guide, it appears that this subject may well be

beyond the grasp of many candidates who would formerly have taken SG maths, as well as

that of many educators, thus raising the question of the country’s capacity to implement

the new syllabus. Maths literacy, on the other hand, will be an elementary subject quite

different to SG maths in the old system. These factors may well dramatically reduce the

numbers of maths graduates who go on to institutions of higher education.

CDE has serious concerns about maths and science education in grades 10 to 12 in the

interregnum before the introduction of FET. Learners proceeding to grade 10 from 2004

onwards will have come through an OBE system based on Curriculum 2005, but will

revert to the old secondary school syllabus until the new FET curriculum starts in grade 10

in 2006. Learners and teachers in this particular secondary school cohort will have to

adjust their expectations of each other in the classroom. This situation will pass over time;

however, since maths or maths literacy will be compulsory subjects in the FET band, the

already inadequate supply of appropriately qualified educators will be overwhelmed by

the new demand. With potentially declining numbers of high-level FETC graduates pro-

ceeding to higher education, the supply of maths educators will dwindle further.

The SC examination

At present, the school-based system of maths and physical science education still culmi-

nates in the SC examination. This will begin to change in 2006 when the new FET curricu-

lum will be introduced in grade 10, and move up year by year thereafter. The first FETC

examination, replacing the SC examination, will take place in 2008.

CDE examined ongoing research in this field, as well as the benchmarking exercise

entered into by the NdoE with the Scottish Education Authority in maths, physical sci-

ence, and other subjects. It then sought the views of those at the ‘coal-face’ on whether

the existing SC examination or marking procedures are creating, or contributing to, the

current low pass rates.

Moderators, examiners, markers, and educators were invited to four workshops to dis-

cuss the SC maths and physical science exams. The workshops confirmed our identifica-

tion of positive elements in our present system. The SC examinations are fulfilling a valid

function, at the correct level, and do not require immediate wholesale intervention. The

Interviewees agreed that

the poor quality and

shortage of educators

was one of the prime

causes of the current

problems

17

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

present system could be improved, when the opportunity presents itself. But, given that

other aspects of the system do require urgent attention, there will be great virtue in keep-

ing the SC exams as stable as possible.

While the participants did have some reservations, it emerged quite clearly that the

level of failures in SC maths and physical science was not attributable to the nature or

standard of the examination itself, its moderation or, with some reservations, avoidable

variations in marking.

CDE has therefore concluded that SC learner assessment is not really part of the prob-

lem, and not an area for wholesale crisis intervention. The present system can be

improved, and curricula could and should be systematically upgraded and modernised..

At the same time, steps should be taken to ensure that future SC exam papers are of the

highest possible standard in terms of coverage, levels of difficulty, and their capacity to dif-

ferentiate between candidates.

However, continuous assessment emerged as an area of concern, and reports that have

reached CDE since the interviews and workshops show a rising level of negative comment.

The principle of non-examination-based assessment can be supported, on grounds of pro-

gressive pedagogy. But how such assessments are made is a key issue. The weighting of 25

per cent of total grade is significant, and can make all the difference between passing and

failing, and one symbol and another. Yet it appears that educators have not been ade-

quately trained in how to administer continuous assessment; that the process is not ade-

quately monitored and controlled; and that provinces do not take the process equally seri-

ously. This area requires urgent attention.

Interviews with educators, education officials, and other experts

CDE also interviewed maths and science educators, officials, and other experts. The inter-

views revealed their belief that the government had a responsibility to take the lead in this

field. They acknowledged that the government had acted to improve the system, and also

that progress had been made. In fact, most believed that too much was being tried at once,

leading to what one interviewee described as ‘policy fatigue’. They felt that fewer initiatives,

centred on the existing strengths of the system, would be more appropriate; that officials,

school administrators, and educators should be given enough time to implement them

properly; and that their outcomes should be assessed before more changes were made.

Respondents unanimously commended the government for introducing the National

Strategy for Mathematics, Physical Science, and Technology. However, they felt that this

important initiative was being undermined by an excess of other education initiatives,

which were not properly co-ordinated or harmonised with maths and science strategies or

programmes. Policies were not properly prioritised, and many had had unexpected nega-

tive consequences. They expressed concern about unco-ordinated changes in curricula,

particularly in the FET band that had, for example, resulted in a situation where learners

who had been exposed to OBE up to grade 9 had to revert back to older-generation curric-

ula in grade 10, as OBE-based curricula and trained educators were not yet available. Simi-

lar comments were made about educator redeployment, and the implementation of con-

tinuous assessment.

Interviewees agreed that the poor quality and shortage of educators was one of the

prime causes of the current problems. Yet all agreed that educators were also central to

resolving those problems. Producing more and better trained educators would take a long

The case studies tell us

that success in maths and

physical science in South

African schools is based

on the most widely

recognised and most

conventional

performance factors

18

F R O M L A G G A R D T O W O R L D C L A S S

time, and it would also be difficult to retain them. The quality of the system could not

change more quickly than this issue could be resolved. Other causes of the current prob-

lems cited were: most practising educators had been educated at low level colleges of edu-

cation and not at universities; the lack of differential salaries to counter the high demand

outside education for graduates with maths and science qualifications; payment by gen-

eral qualification and years of experience only; poor working conditions; a lack of learn-

ing resources for maths and physical science; and pressures to produce ‘matric results’. A

more flexible and innovative policy was needed that would allow school governing bodies

to offer maths and science educators additional incentives, without adding to general

expenditure on personnel.

Finally, respondents were asked what practical steps could be taken to improve maths

and physical science education. At the provincial department and school district level,

they identified: improved monitoring and support by curriculum advisors and officials;

appropriate in-service courses (preferably focused on content knowledge); and a more per-

sonalised, targeted, differentiated, and less bureaucratic approach by officials. At the

school level, they identified: strong leadership; good management; a results-oriented cul-

ture; reasonable class sizes; confident educators; a supportive environment for learners;

better and earlier diagnostic assessment; and strong parental and community support. In

the classroom, they identified: the effective planning and pacing of work; stressing the rel-

evance of both subjects to daily life; more personalised contact with learners; and provid-

ing learners with additional learning materials and courses.

In the course of these interviews, CDE detected a further virtue in the present situation:

the high level of concern among educational experts that South Africa should do better,

and the conviction that it can do better. Appropriate initiatives – perhaps launched as a

result of this report – will certainly find a sympathetic audience among these practitioners

and their peers.

Case studies

Perhaps the most significant qualitative perspective was that which emerged from a study

of 13 schools in five provinces whose performance in maths and physical science was

either excellent or improving, relative to comparable schools. The selected schools were

• Tsogo High School : Located in North West, in a peri-

urban, ex-bantustan area 45 kilometres west of Pretoria.

Some 65 per cent of learners from townships, and 35 per

cent from rural and informal settlements. Cumulative

maths performance, 1996-2001: candidates – 405; HG

passes – 146; SG passes – 253: pass rate – 98,5 per cent

• Ahu Thuto Secondary School: Located in Orange

Farm, Gauteng, a vast informal settlement 50 kilometres

south of Johannesburg. The area is marked by poverty

and unemployment. Some 85 per cent of learners from

Excellent schools in unexpected places

the surrounding townships, and 20 per cent from the

informal settlement. Cumulative maths performance,

1996-2001: candidates – 263; HG passes – 10; SG

passes – 210; pass rate – 81 per cent.

• Siyangempumelelo High School: Located in rural

KwaZulu-Natal, in a former tribal reserve, 34 kilometers

from Nongoma. All the learners come from the

surrounding rural areas. Cumulative maths performance,

1996-2001: candidates – 160; HG passes – 28; SG

passes – 84; pass rate – 68 per cent. CDE 2004

A remarkable coherence

of views emerges from

our qualitative research

regarding the scale and

nature of the problem,

and the actions that

should be taken

19

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

diverse in terms of the socio-economic backgrounds of learners, resources, history, and

other factors. All they had in common were a principal, some classrooms, some educators,

and a majority of African first-language learners. Seven of the schools consistently per-

formed far better than the national average in maths and physical science, while the

remaining six were performing better than they had in the past, and somewhat better

than the national average. All but two were ordinary African public schools in five differ-

ent provinces, located in rural KwaZulu; urban townships, including Khayalitsha and

Soweto; informal,’peri-urban’ settlements including Orange Farm; and former bantustan

areas, including Hammanskraal.

The case studies tell us unequivocally that success in maths and physical science in

South African schools is based on the most widely recognised and most conventional per-

formance factors, namely the quality of the principal; the competence of the educators;

the effectiveness and efficiency of school management and administration; adequate dis-

cipline in all aspects of school life; healthy competitiveness within the school and among

schools; and relatively traditional methods in the classroom.

Obviously, resource and other inequities are evident across the schooling system, and

play a large part in determining learners’ chances of success in maths and physical sci-

ence. However, remarkable – even miraculous – virtue was often found where such nega-

tive socio-economic factors weighed most heavily.

The educator is at the centre of all efforts to improve maths and physical science

learner performance, including so-called ‘learner-centred’ methods, which are based on

an educator’s decision to use this approach. Successful educators emphasise clarity of

explanation, repetition of problems, volume of homework done and marked, feedback on

performance, and specific preparation for examinations once the syllabus has been cov-

ered. Good relationships among the school, parents, and the surrounding community also

contribute to school and learner performance.

Effective teaching is occurring in very inauspicious settings. However, this virtue is not

as widespread and consistent as it could be. A small percentage of South African schools

continue to supply a high percentage of the successes. We also encountered some illumi-

nating faults, both of omission (ie, not doing well-known things), and commission (ie,

doing things that were known to be incorrect).

On balance, though, we can now state with confidence what the virtues are within

schools to be built on or created, what the faults are, and also what practical steps can be

taken to remedy the situation. Many basic reforms are still necessary in many extremely

underresourced schools, and should be pursued energetically to the point where all sec-

ondary schools possess an essential minimum of educational resources.

Conclusions from the qualitative research

A remarkable coherence of views emerges from our qualitative research regarding the

scale and nature of the problem, and the actions that should be taken. The interviews,

workshops, and case studies support each other in identifying specific ‘virtues’ that repre-

sent opportunities on which to build successful reforms. Interviewees were also united in

their criticism, and even in their despair over the present system.

This unanimity of views also represents an opportunity. It means that there is a group

of committed professional educators in the country whose views are consistent and could

be galvanised into action by good leadership and well-planned initiatives. This group also

We now have sufficiently

detailed data about the

system to allow us to

intervene locally in

different ways, within a

consistent overarching

policy and strategy

20

F R O M L A G G A R D T O W O R L D C L A S S

understands that the problems cannot all be overcome at the same time, or by means of a

single policy initiative, and that the policies applied in schools need to be sufficiently flexi-

ble and pragmatic for the virtues already evident in the school system to be built on. In

addition, ample information now exists about the views of different groupings within the

body of concerned professionals, which can be the basis of a sophisticated campaign to

gain their support.

The principals, parents, educators, and learners interviewed by CDE are knowledgeable

about the factors of success, and about the actions necessary for excellent maths and

physical science performance. Instead of planning system-wide and undifferentiated

reform policies, education theorists, policy-makers, and departmental officials should be

‘looking for virtues’ in the existing system. These ‘virtues’ clearly demonstrate that the

system is already capable of a much better performance, provided the basics are attended

to and improved.

INTERNATIONAL EXPERIENCE

Flowing from our international research, we believe South Africa is capable of launching

an effective strategy to improve the ‘quantity of quality’ maths and science HG passes. A

high level of conceptual understanding has been developed internationally, and success-

ful practical interventions have been made in many national systems. These have been

exhaustively evaluated and documented. It remains for us to apply these findings intelli-

gently in this country. Some of the key lessons distilled from our survey of international

research, and their local implications, are:

Systemic change is the only approach likely to succeed in a large educational

system that is in active operation and has deep-seated problems.

There is now almost universal acceptance internationally of the concept of ‘systemic

change’ or ‘systemic reform’. Much has been written about this concept, some of it aca-

demic in nature, and difficult to use in real systems. Some writers appear to confuse ‘sys-

temic’ with ‘radical’, and assume that systemic change involves rapid fundamental change

of an entire system. This is misleading, as systemic change is essentially a conservative

approach, in the sense that it acknowledges that a working system already exists, and seeks

to conserve what is good in the system and build on it, rather than starting afresh.

Small systems might, in principle, be changed all at once, while systems that are no

longer functioning can be reconstructed from scratch. However, South Africa’s maths and

science education system is operating in many dimensions with a measure of success, it is

large, and must be kept going while improvements are made. It is essential to balance

change with the need to maintain the best possible quality in routine operations. Only a

portion of available resources can be committed to reform; most have to remain committed

to routine operations. This is another compelling reason for careful, step-by-step change.

We must think holistically, but act incrementally.

Our understanding of systemic change means that we need to plan holistically to act at

appropriate local levels, and in respect of local components of the system. There are obvi-

ous criteria for selecting points at which to act: the change must be manageable; goals

The South African school

population has at least as

many potentially

successful maths and

science learners as any

other national system

21

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

djfksdjdfkjk

The ‘102 dedicated maths and science high schools’ pro-

gramme – subsequently renamed Dinaledi (‘stars’) – was

launched in 2001 as one of the key components of the

National Maths, Science and Technology Strategy (NMSTE).

In the words of the NMSTE annual report for 2003, it is

aimed at increasing the participation and performance of

historically disadvantaged learners in 102 schools distrib-

uted ‘pro rata across the nine provinces’. The report also

offers an assessment of Dinaledi three years after its com-

mencement. A brief summary follows.

Achievements

• Raising maths and science participation and performance

in the 102 dedicated maths and science schools.

• A ‘steady increase’ in enrolment and pass rates.

• ‘Encouraging’ improvements in African maths and

science participation and performance.

Constraints

• Poor output of maths and science graduates in grade 12

– particularly at HG.

• Underqualified and unqualified maths and science

teachers, feeding into a ‘vicious cycle of low-quality

teaching, poor learner performance, and a constant

The Dinaledi Programme: NdoE three-year review raises tough questions

undersupply of quality teachers’.

• A lack of adequate facilities and resources for effective

teaching and learning

• Insufficient financial and other forms of support for

talented students

The new minister of education, Naledi Pandor, has stated:

‘We will consolidate the efforts made thus far in improving

the teaching of mathematics, science, and technology in our

schools. Not only will we continue supporting the 102

Mathematics and Science focus schools, we will provide the

resources necessary for the proper teaching and learning of

these subjects in many more schools, both at primary and

secondary level.’

The Dinaledi programme is clearly a vital initiative. How-

ever, after three years of operation, some important questions

arise. How should specialist schools be selected? Is there a

process for additional schools to qualify in time? Exactly what

is being achieved in each school? Which aspects of the pro-

gramme are working well, and which have not turned out as

expected? Do we have the capacity to expand the pro-

gramme to many more schools, bearing in mind the NDoE’s

own finding about the shortage and lack of skills of existing

educators? In the final section of this report, we make a spe-

cific recommendation about Dinaledi. CDE 2004

must be realistic; outcomes must be visible and demonstrable; components of the system

must be selected that are functioning quite well but could rapidly be improved to become

excellent; others must be failing so visibly that they are undercutting any positive out-

comes; and so on. The criteria for action must be different in different cases, because the

system itself is heterogeneous. However, as we have seen before, we now have sufficiently

detailed data about the system to allow us to intervene locally in different ways, within a

consistent overarching policy and strategy.

We must intervene at the appropriate level in order to create positive outcomes

as soon as possible.

International literature makes it clear that systemic change works optimally by identify-

ing basic building blocks upon which positive reforms can progressively be attached. It is

important to create positive outcomes as rapidly as possible, as these create a sense of

progress, and motivate participants. Some interventions need to be considered in the

framework of long-term goals. Others have to be considered in terms of quick gains,

strengthening motivation, providing models that can be dispersed more widely, and other

strategic objectives.

Despite energetic

government efforts over

the past 10 years,

significant results have

not been achieved, and

disillusion is setting in

22

F R O M L A G G A R D T O W O R L D C L A S S

A successful way of achieving focus is to group efforts relating to a particular problem

into a ‘programme’, with specific participant inputs, activities, and goals that can develop

quickly and provide results that can be generalised to the whole system. A restructured

Dinaledi Programme could play this role, leading to the more widespread acceptance of

specialist schools.

People come before structures.

The priority area for initiating systemic change is not structures, but people. This is partic-

ularly true of education systems. The key factor in any systemic change is the supply and

retention of competent and confident educators, closely followed by the supply of effective

and efficient principals.

The quantitative studies commissioned by CDE show clearly that the South African

school population has at least as many potentially successful maths and science learners

as any other national system. Our immediate task is to ensure that more learners enrol for

SC maths and science, and pass these subjects at the HG (or, from 2008 onwards, its FETC

equivalent).

Added to this, we need to attract appropriate numbers of school-leavers to careers as

educators in maths or physical science, and also improve the content knowledge and

teaching skills of existing educators. There is little doubt that this is where our pro-

gramme of systemic reform should begin. Using a variety of incentives, we must attract

school-leavers to higher education programmes that will lead to their employment as edu-

cators. At the same time, we must retain and improve all existing educators. Using differ-

ent incentives, we must attract some of those who have left back into the system. It is an

illusion to think that these goals can be achieved if we continue to treat maths and science

educators in the same way as all others.

Every study that CDE is aware of shows that educators in developing countries tend to

lack confidence, and become fearful and demotivated when confronted with large-scale

changes over short periods of time. As a consequence, successful programmes avoid

changing too many features at once. It would be inappropriate to change curricula, teach-

ing methods, and examinations simultaneously, however strongly we might feel that these

three components should dovetail. Dovetailing can be achieved over time; an improved

curriculum with unchanged teaching methods and the same exam format would in itself

represent an advance. The low confidence levels of educators also lead to two other obser-

vations. The proposed changes must be canvassed thoroughly with educators, and effec-

tive training provided in respect of changed procedures. Also, the main source of the lack

of confidence of most educators is poor content knowledge, and this issue must be

addressed before attempts to change teaching methods are made.

People respond to incentives

Successful systemic change is underpinned by a single fundamental truth: development

depends on people, and people respond to incentives. Looked at from this perspective, an

educational system is nothing more or less than a complex system of incentives that pro-

vides people with rewards for doing what they wish to do to the best of their ability. Such

incentives are by no means limited to financial rewards, but also include opportunities for

personal and professional development, peer recognition, praise from superiors, respect

Hundreds of thousands of

young South Africans are

enrolled in schools where

they have little or no

chance of passing HG

maths and science

23

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

from subordinates, and many other factors. Systemic reform, then, should not start with

the question: ‘How can we make people do this?” It should start with the question: ‘What

incentives can we offer people to do this, and become motivated to do it better?’

POINTS OF DEPARTURE

The insights gained from our research have been used to formulate 12 key propositions

which will in turn serve as points of departure towards our practical recommendations in

the final section.

1. Maths and science are crucial to South Africa’s success.

National development requires an increasing number of skilled personnel. Specifically,

they require competencies based on mathematical and/or scientific knowledge.

Despite energetic government efforts over the past 10 years, which are described and

analysed in depth in our main report, significant results have not been achieved, and disil-

lusion is setting in. This emerges clearly from our interviews with educators and educa-

tion officials, as well as our school studies. These feelings contribute to a perception of

national inability to stimulate development and meet promises, and will worsen should

maths and science education be perceived as an area where delivery fails to match reason-

able expectations.

We must and can persuade the nation as a whole to accept the significance of maths

and science reform, and participate in it. A wide variety of groupings have legitimate and

positive roles to play. We must go beyond a sense that maths and science education is sim-

ply ‘the government’s job’, and mobilise resources on a nationwide basis.

2. Failure with respect to maths and science education is the most important

obstacle to African advancement.

CDE’s research has clearly demonstrated the extent of the national crisis in maths and science

performance and participation since 1991. The effects of this national crisis are most evident

in respect of African learners. To restate our findings: in 2002 only 4 637 African learners

passed HG maths, representing only 13,14 per cent of all SC candidates, and only 23,42 per

cent of all HG passes. And only 7 129 African learners passed HG science, representing only

14,06 per cent of the total number of SC passes, and 30,42 per cent of all HG passes.

This is holding back African advancement. It places a huge obstacle in the way of

achieving almost all the government’s ambitions to open up vast new areas of opportu-

nity for black South Africans. The private sector’s efforts to apply government policy and

open its doors to Africans are held back if there are insufficient numbers of qualified can-

didates for increasingly skilled positions. Most of the jobs now being created require com-

petence in at least mathematics – this includes work at nearly all skills levels, in the manu-

facturing, construction, retail, service industry, engineering, and technology sectors.

Successful businesses, enterprise creation, and management skills across the board

require a competence in maths and languages. Access to higher education and the profes-

sions is in almost every case precluded without, at least, maths qualifications.

If we want to ensure African economic empowerment, increased employment equity,

We do not have the

educator capacity to

reform the system

immediately, even in

grades 10–12. We still

have to build that

capacity

24

F R O M L A G G A R D T O W O R L D C L A S S

and growing numbers of Africans in more senior positions in the economy and society, we

have to dramatically improve the number of school graduates in HG maths and science.

This will also require more qualified candidates entering the education profession, and

becoming dedicated and effective teachers in these subjects. There can be few higher pri-

orities in South Africa today.

3. The present maths and science education system is failing many individuals

and their communities, and is wasting national resources

Hundreds of thousands of young South Africans are enrolled in schools where they have

little or no chance of passing HG maths and science. Simultaneously, there are thousands

of young learners who, if given the right guidance, as well as competent teachers, would

pass SC maths and science in the HG.

This is key to all our ambitions concerning community upliftment and local economic

development not to mention individual or family advancement. An example from Gauteng

illustrates this point: in 2003 the two Ekhuruleni townships of Tsakane and KwaThema,

situated at the heart of the national economy, achieved 1 600 SC passes – but only 12 of

these included HG maths.

4. We need to acknowledge the diversity and complexity of the system, and

support targeted interventions over single policy approaches.

South Africa’s schooling system is very large, and there are important variations in the

conditions under which maths and science education are provided. One single, central,

national policy and strategy for improving maths and science education would be inap-

propriate. Rather, within a broad framework, focused policies are required aimed at

achieving specific outcomes in specific settings, utilising different strategies and methods

of implementation tailored to individual schools and their particular levels of teaching

skills and learner awareness.

A major obstacle to differentiated policies and implementation has always been inade-

quate information about the subcomponents of the system, whether these be geographi-

cal, managerial, functional, or learner-related. This project has started the process of

overcoming this obstacle. We now have reliable information on performance in the SC

examination over time, and in each of the provinces; the status of individual schools by

name, location, and facilities; a profile of individual school performance year-on-year, and

in comparison to all other schools; and individual learners, even the specifications of

groups whose members could pass either or both subjects at HG but do not enrol for these

subjects.

Therefore, it is now perfectly possible to act in different ways in respect of different parts

of the system. But the reasoning and logic for this graduated and phased approach needs

to be accepted by government, and communicated to all concerned, so that no misunder-

standings develop about the government’s medium- and long-term commitment to equal

access to quality maths and physical science learning and teaching for all.

When such an approach is adopted, issues arise around the balance between ‘transfor-

mation’ and ‘diversity’, and ‘equity’ and ‘excellence’. But that is exactly the point. This

area of policy requires an ongoing debate among stakeholders within the general princi-

ple of differentiated policies, implemented intelligently in different ways in different parts

South Africa does not yet

have the resources to

introduce a bottom-up

reform initiative, starting

in primary schools

25

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

of the system, according to local conditions. This will involve a wider spread of responsi-

bility that will, in turn, release much greater initiative and responsiveness in the system.

5. We cannot change everything at once. Priorities are required.

Research cautions us to be realistic. Despite all the ‘virtues’ we have identified, we still

haven’t reached square one. The reality is that we do not have enough educators compe-

tent to teach maths and science, nor sufficient learners competent enough to study these

subjects at the levels we require and in numbers that will make a difference to the econ-

omy and the education system. We certainly do not yet have the resources to introduce a

bottom-up reform initiative, starting in the primary schools.

The implication of this is that, before we proceed, we must agree on the key priorities

for immediate action. The system is too heavily constrained to attempt many things at

once. In a situation of limited capacity in government, provinces, schools, the educator

community, and the private sector, only a few priority areas can be addressed, but these

must be the key priorities.

Our research indicates that we must intervene in grades 10 to 12, keeping as much of

the present system intact as possible, and focus on increasing the number of good HG

passes over a short period. Then we must try to draw as many of these graduates as possi-

ble back into education, and retain them by any means necessary, including incentives

appropriate to their individual and professional needs. At the same time, the content

knowledge and teaching skills of existing maths and science educators must be improved

and upgraded. Vigorous attempts must be made to re-recruit well-qualified and skilled

educators who have left the system. We must also accept the vital role played by language

competence, and launch appropriate programmes to improve educators’ skills in the lan-

guages of instruction and examination. We must be clear that, while this is only a first

phase, it is a vital one without which further advances will not be possible.

Every resource available to us will be needed to achieve only these few priorities for

action. We do not have the educator capacity to reform the system immediately, even in

• Increase the number of qualified maths and science

educators.

• Ensure that each educator has a sound knowledge of the

curriculum.

• Improve the professional development of educators.

• Upgrade the language competencies of maths and

science educators and learners.

• Identify learners with an aptitude for maths and science

early in the system.

• Give more time to maths and science in the school year.

• Benchmark performance, domestically and internationally.

• Apply policies and programmes in a flexible manner,

relating them to the needs of individual schools.

Maths and science challenges in brief

• Launch programmes specifically aimed at improving pass

rates in maths and science, including specialist schools,

incentives for educators, and financial assistance to

learners.

• Act on critical factors identified in the school-based

research undertaken by CDE and others.

• Strengthen school management and administration.

• Strengthen school governing bodies and allow them to

take substantive decisions in respect of maths and science.

• Temporarily suspend the maths and science programmes

in extremely weak schools, and redeploy qualified

educators and talented learners to stronger schools.

CDE 2004

A sound theoretical basis

exists for using incentives

to improve the

recruitment, development,

and retention of maths

and science educators

26

F R O M L A G G A R D T O W O R L D C L A S S

grades 10-12. We still have to build that capacity. Policies and strategies cannot assume

that we have the capacity even to train our present corps of educators to deal with new

curricula and teaching methods.

6. Additional properly qualified and trained maths and science educators must

be found

There is no reliable national database on the qualifications of educators in South African

schools. The best information available is based on self-reporting by educators concerning

their qualifications. By this measure, only 14 per cent of schools reported in 2001 that

their maths and science educators had the minimum qualifications prescribed by the

NdoE (SC plus 3,5 years of higher education). The country faces a severe shortage of

trained, qualified, and experienced educators if we wish to expand participation and

improve the performance of learners at the levels required.

Experts indicate that the norms for higher grade maths and science teaching need to be

higher than at present. For maths, the norm should be that all teachers who teach HG

maths (or maths as opposed to maths literacy in the proposed new system) should have a

university degree with a mathematics major. For science, all teachers should have physics

and chemistry majors in their university degree or one major with the other at second

year level.

As temporary measures one could use people with Maths 11 for maths or people with

two years in physics and chemistry for science. In general three or four year college diplo-

mas will not do for higher grade teaching in grades 10-12. Finding out who these teach-

ers are in our system will take a special audit. Once we have this audit we will be in a posi-

tion to measure the true extent of the enormous educator challenge that the country

faces.

Serious consideration needs to be given to retaining our many excellent teachers (many

of whom work under difficult circumstances); utilising their skills as effectively as possible,

so that as many learners as possible can benefit from their expertise; attracting more of our

small numbers of appropriately qualified SC graduates back to the teaching profession;

ensuring the best possible system of upgrading the knowledge and skills of those who are

already in place (CDE has identified international programmes that are directly relevant);

and, if necessary once we know the results of a detailed audit of current educator qualifi-

cations, hiring qualified and experienced English-speaking educators from other countries

(India, parts of eastern Europe, and other African countries offer real possibilities).

7. Anyone with aptitude and initiative should have access to maths and science

education

The maths and science education system must be purposefully broadened to encompass

every learner with aptitude and the initiative to take up the learning challenge involved.

However, not every learner has the capacity or interest to succeed at the highest levels,

nor to go on to a degree and an education qualification. Therefore, the value of access

across the system must be pursued at the same time as excellence is achieved in specific

parts in the system. Mechanisms must be found to ensure that ‘no child with potential in

maths and science is left behind’ because they live in a part of the country that has been

historically neglected. This will require learners with potential to be identified at an early

Private education is

growing rapidly in most

developing countries. It is

no longer synonymous

with ‘elite’ education

27

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

stage; and the country providing the organisation and resources needed to enrol them in

schools competent to teach maths and science.

8. Incentives play a decisive role in development.

Development literature, evidence of what works internationally, and numerous projects

and programmes increasingly show that incentives play a decisive role in human develop-

ment, and thus in social development. It has become clear that development is often

unsuccessful if it simply exhorts or encourages individuals, or even seeks to create specific

opportunities. Meaningful incentives also need to be provided. These must be measurable

advantages to individuals that, if taken up, will assist both the individual and society.

Incentives are already a major feature of the modern workplace: these arrangements

are now so prevalent that an influential commentator, Professor William Easterly has

stated: ‘People respond to incentives: all the rest is commentary.’

Our research shows that a sound theoretical basis exists for using incentives to improve

the recruitment, development, and retention of maths and science educators. They need

to be carefully implemented; however, if applied sensibly, incentives can play a positive role

in overcoming constraints in the supply of and demand for maths and science educators.

They can also encourage more students to study HG maths and science when they see the

rewards available to them for working at these ‘difficult subjects’. Lastly, they can also be

used to encourage schools to improve their maths and science performance.

9. State and markets – supply and demand in education.

In country after country in the developed and developing world, progressive modern

states are providing the private sector increased scope and opportunities as the providers

of more differentiated, specialised educational opportunities, often in response to rising

demand from parents.

This is being accomplished in countries with both large and small populations, and

without governments relinquishing their obligations for continuing to ensure that an effi-

cient and effective public system of education exists. By creating the space for private

providers in education, many public education authorities have successfully combined a

degree of deregulation of the educational sector with expanded opportunities for private

education providers to operate in tandem and/or parallel with public education. Increas-

ingly, public monies follow the learner – whether that learner attends a registered public

or independent school. As long as the regulatory system ensures public accountability for

performance and some aspects of the curricula (a commitment to support the country’s

democratic constitution, for example), such systems have positive outcomes on perform-

ance. A compelling example is the provision of publicly funded school vouchers for use by

poor inner-city students, often drawn from minority communities, in the United States,

and supported by a majority of African-American parents nation-wide. Another example

from countries in the EU and the US involves public schools operating under special agree-

ments with state education authorities to provide specialised education – so-called ‘char-

ter schools’.

Private education is also growing rapidly in most developing countries, and, as a recent

International Finance Corporation report shows, it is no longer synonymous with ‘elite’

education. In fact, the opposite effect is evident in that many poor learners previously

‘Fixing’ South Africa’s

system of maths and

science education is

the country’s top

educational priority

28

F R O M L A G G A R D T O W O R L D C L A S S

trapped in underperforming, underresourced, poorly managed public systems now have

access to reasonably priced, good-quality, and standardised educational opportunities pro-

vided by global or local education companies. South Africa is not excluded from these

global processes. Our research shows that stimulating demand within the publicly pro-

vided system can also encourage greater flexibility and efficiency of provision; improve

the motivation of principals, educators, and learners; and generally add value to the sys-

tem.

The 21st-century state has an obligation to provide a public system of education and

should do so as efficiently and effectively as possible. However, there is no value in insist-

ing that public provision and overall accountability for a quality education in a given sub-

ject precludes working with private providers of education, or responding to demands

from parents and learners for the more flexible use of public resources.

Real advantages accrue to countries that encourage a thousand (educational) ‘points of

light’ to burn, while keeping a stern eye on quality control and standards. Flexibility of

provision and responsiveness to need must balance the acceptance of responsibility by the

public sector.

10. We need to demonstrate new attitudes towards educational reform.

The success of the initiatives CDE is proposing will depend on pragmatism, flexibility, pri-

oritising the needs of one part of the system over others (at least temporarily), and provid-

ing special incentives for and status to certain educators and learners. The country needs

to approach elements of educational reform in a more pragmatic and nuanced way. Such

an approach in no way diminishes the long-term goals of equity and excellence, but does

mean more flexibility in how we achieve these goals. Some areas of change must be priori-

tised over others. Excellence must be seen as the precursor of and not the successor to

equity. Equity of provision in maths and science can only be achieved through excellent

principals and educators. There may even be a need to discontinue expenditure on maths

and science in long-standing unproductive areas, such as schools with long histories of

pass rates below 20 per cent, so that we can focus all resources where the most positive

results will be achieved most quickly.

Obviously, the approach CDE is proposing differs from any of those used thus far. We

believe this is necessary and beneficial. Too many interventions have been attempted

without lasting impact, and it is time to try something new. But we need to try it in a disci-

plined, conservative, and incremental way, in order to minimise possible negative side-

effects, and maximise our chances of success.

11. The importance of a monitoring and research base for action.

Proposals to change even a badly functioning system have to be based on sound research

of that system, and located in a solid understanding of what is being tried elsewhere.

We need to identify and monitor key indices of improvement. The system itself cannot

be improved without frequent assessments of how it is performing, where its strengths lie,

and which schools continue to disappoint despite the application of resources and sup-

port. We must, however, agree on what these key indices are: in our view, it is the increas-

ing quantity of quality passes.

South Africa has erred by moving away from system assessments, and probably has too

Any attempt to improve

the maths and science

education system must

start with limited but

achievable aims that

will lay the foundation

for an improving

system over time

29

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

few individual examinations as well. We need to sample our system more frequently,

assess patterns of performance and participation in our schools, and begin to target our

resources within a clear programme that supports participating schools and rewards indi-

vidual schools on their respective merits.

12. The government must lead a national partnership that delivers results.

As in any large-scale system, government must provide a clear framework for action, with

appropriate guidelines and targets for implementation at different levels.

However, this programme must be drawn up in consultation with all role players, and

the government must ensure that its goals and outcomes are consistent with all its other

actions and policies, and that individual school communities are empowered to respond

positively to these initiatives. In the last instance the responsibility does lie with govern-

ment to introduce and drive such a programme via an appropriate body, but it must find

positive and productive ways of working with other role players.

Conclusion

‘Fixing’ South Africa’s system of maths and science education is one of the country’s

most important national priorities. Energy, goodwill, and capacity to move ahead is evi-

dent across the public and private sectors. What is needed is an achievable set of activities

that will demonstrate rapid progress, and improve confidence in our ability as a country to

turn things around. Leadership is required that understands the constraints within which

we need to move forward, and can pull together the energy and resources of all those will-

ing and able to make a difference. We will now make practical proposals on how to move

forward.

An incentives programme

for maths and science

educators should be

launched immediately. It

must be aimed at

attracting new educators,

as well as retaining the

skilled educators we have

30

F R O M L A G G A R D T O W O R L D C L A S S

CDE’S RECOMMENDATIONS

South Africa has to deal with a national crisis with respect to maths and science school-

ing. Current public and private sector efforts are insufficient to significantly change the

system that is failing individuals, families, communities and the country. Almost all South

Africa’s ambitions to grow the economy, provide new job opportunities for black citizens,

and ensure our success as a democracy are undermined by our collective failure in this

area. We do not have the capacity to change everything we would like to immediately. A

limited programme of action needs to be decided upon. Dramatic improvements in the

number of quality passes are possible in the short term. Achieving these will require a

new and common framework of understanding and investment by public and private

leadership and other key players.

In this final section, we will define goals for a new approach to maths and science edu-

cation in South Africa; outline our overall approach to reform; and put forward ten practi-

cal proposals.

Goals

Over the next five years, South Africa should aim to:

• double the number of HG SC maths and science passes

• double the number of qualified and able teachers in the public school system

Approach

CDE’s recommendations are based on the following underlying guidelines that have

emerged from our research:

• Any attempt to improve the maths and science education system must start with lim-

ited but achievable aims that will lay the foundation for an improving system over time.

• New initiatives must build on what is working in the system, ensure that performing

schools continue to excel while changes are introduced, and ensure that their ability to

deliver quality education is available to as many individuals as possible.

• During the next five years, every learner with aptitude and initiative, wherever he or

she may live, must have the opportunity to study at a school that provides effective

maths and science teaching;

• There is no “single best national way” to improve maths and science education through

centralised policies.

• Public and private sector leadership, energy, and initiative must pull together if we are

to succeed.

Proposals

We must introduce a comprehensive programme that provides for systemic change,

involves public and private sector leadership and resources, achieves short-term results,

and is implemented via an effective and accountable institution.

Steps should be taken to

identify all high-

performing schools and

investigate ways in which

they could play an even

bigger role

31

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

Mobilise the concerns of important stakeholders in mathsand science education into a national programme.

There are many stakeholders in maths and science education, including national and

provincial government, the private sector, foreign governments and donor bodies, interna-

tional agencies, the independent schooling system, tertiary education, the scientific

research community, educators, learners, and, last but not least, parents.

There is a worldwide recognition that this is a difficult field, and few countries are satis-

fied with their performance. Therefore, the challenge must be seen as a long-term one.

Improving and reforming the system must be regarded as an ongoing process, not a single

goal to be reached at a specific point in the future. We have to plan to produce a better sys-

tem with a built-in capacity to improve further by monitoring and evaluating its own per-

formance, and making additional incremental changes. In order to do this, we must assess

progress much more often; and test learners more frequently. The outcome we all seek is

better results, and a gradually improving system. In the process, we should aim at creat-

ing greater public confidence in the education system as a key pillar of a successful and

increasingly prosperous democratic society.

Mobilising the energies and commitment of all the relevant role players will require an

energetic but focused programme. We need a systematic, country-wide initiative aimed at

ensuring a common understanding of the nature of the challenge we face and mobilising

commitment and involvement towards the common strategic set of interventions pro-

posed here.

The key to successful reform is an increased supply of quali-fied maths and science educators.

Effective educators are vital to improving the system, but they are in short supply, thus con-

straining our reform efforts. Three programmes should be devised to deal with this situation:

• Identify the qualified and able maths and science educators currently in the schooling system.

No one has reliable and comprehensive national information. This is urgently needed

for the programmes we are suggesting, and also because our research indicates that

there are a large number of educators qualified to teach maths and science but who are

employed to teach other subjects.

• Increase the supply of qualified maths and science educators, and retain existing competent

educators by means of a well-conceptualised programme of incentives. Maths and science

educators are in short supply throughout the world, and graduates with appropriate

qualifications are at a special premium in South Africa. Nevertheless, the issue of provid-

ing incentives to educators in certain subjects has long been resisted here. An incentives

programme for maths and science educators should be launched immediately. It must

be aimed at attracting new educators, as well as retaining the skilled educators we

have. The best results will probably be achieved if the innovations are introduced as a

distinctive ‘programme’ and not as part of the routine public administration of this sec-

tor. This will also give the programme a better chance of attracting financial support

from other institutions, including private sector corporations, as it will give their fund-

ing a specific focus.

• Institute a new approach to the professional development of maths and science educators; if

2

1

All maths and science

educational activities

should be closely linked

with improved language

education

32

F R O M L A G G A R D T O W O R L D C L A S S

necessary, adopt successful models from abroad. The first emphasis in developing educa-

tors must be on content knowledge, followed by teaching skills. This is the basis of the

most successful educator development programmes CDE has identified in the course of

its research. Their success is also based on the principle of steady incremental improve-

ment achieved within a comprehensive programme of support, marked by regular

assessments, combined with appropriate incentives to participants. The motivational

and communications dimensions of these programmes are also impressive. Educators

who benefit from these programmes are expected to pass on some of their new-found

knowledge and skills to colleagues who have not or could not attend. Should a high-

profile international programme be adopted, other national government aid institu-

tions, corporations, and private foundations might be persuaded to financially support

its introduction in South Africa.

Build on the potential in the school system.

Given CDE’s school performance index, we are now able to identify and classify the maths

and science performance of every secondary school in the country. This enables us to for-

mulate specific initiatives to improve their performance, and the access of talented learn-

ers to well-performing schools. Steps should be taken to:

• Identify all high-performing schools (ie schools with a pass rate of 80 per cent or more in

large HG maths and science classes), and investigate ways in which they could play an even

bigger role. Can they deal with larger classes; can they expand their maths and science

departments; can they share their expertise with poorer performing schools in their

neighbourhoods; can a city run a programme to encourage performing schools to

‘adopt’ the maths and science departments of non-performing schools in other parts of

the city? Appropriate incentives must be provided for these schools to play such an

expanded role.

• Devise specific programmes to help those schools delivering pass rates in the 60-80 per cent

band to improve their performance. These could include programmes to improve general

school management, and programmes to improve the skills of educators teaching

maths, science, and the languages of instruction in those subjects. Again, incentives

should be provided to motivate school administrators and educators to move to a

higher performance bracket.

• Establish an independent, objective measure of each and every school’s annual performance in

maths and science. Incrementally ‘raise the bar’ for every participating school by setting

realistic targets for increasing participation and performance. Link incentives to goals,

so that schools are encouraged to progressively climb the ladder of success with respect

to maths and science delivery.

‘No child left behind’: provide mechanisms for learners andparents, wherever they live, to take advantage of neweducational opportunities.

South Africa’s educational system has to deal with the legacy of apartheid. This means

that many poorer, mainly African, households in urban as well as rural areas do not have

3

4

Concerned private sector

funders should use this

report as the basis for a

discussion on how to

focus their input more

strategically

33

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

access to good schools with functioning maths and science departments. This harsh real-

ity cannot be fixed overnight, and CDE’s proposals are designed to incrementally increase

the country’s supply of decent educational opportunities for everyone. This will take time.

In the meantime, there are initiatives we can undertake to ensure that no one with initia-

tive and aptitude need be denied opportunity.

We need to find ways of stimulating greater demand by parents and learners (and dedi-

cated educators and principals) for quality maths and science education, by providing

new avenues for accessing opportunity in maths and science.

A trial programme should be launched that works as follows:

• The introduction of a national aptitude test, available throughout the country on an

annual basis. This should be for grade 9 learners and should be independently set,

marked, and monitored.

• Any learner who does well in this test should be eligible for financial support to attend a

school with a good delivery record in SC maths and science. This could be a neighbour-

ing school, when all that would be needed would be a transport subsidy; or they might

need to go to a boarding school, in which case more resources would be required.

• Money will follow learners – in other words, a learner would take his/her allotted pub-

lic subsidy with him/her to the new school, which would, in turn, benefit educationally

and financially.

• A pool of new funds will be needed for additional costs: running the aptitude tests

around the country, providing boarding fees or transportation costs or both for promis-

ing learners.

This is a new proposal for the redistribution of financial resources towards poor parents,

and will need to be thought through carefully, and implemented experimentally. But its

long-term beneficial impact seems clear. Demand for better maths and science education

will have a positive impact on the entire system, as schools lose or gain learners, and will

provide incentives to schools to change and improve. There is international experience on

which to draw. This would be an ideal area for attracting private sector support and

involvement.

If we want results in the next five years we need to get much better matching between

good learners, good educators and effective schools. This set of proposals – an aptitude test

to identify learners with potential and then the mechanisms and resources to get those

learners to effective schools who teach maths and science at HG level properly - will help

the country to do this and should be implemented as a matter of urgency.

All maths and science education initiatives should includeappropriate language components.

For ease of reference, CDE has used the phrase ‘maths and science education’ throughout

this and the main report. However, as we have noted earlier, learners’ proficiency in the

language of instruction and examination plays a very significant role in their perform-

ance in maths and science. This has been confirmed by CDE’s case studies, our interviews

with practitioners, and our workshops with examiners. It is also stressed in the interna-

tional research.

As a result, all maths and science activities should be closely linked with improved lan-

guage education. Given the nature of global economic development, it will be most benefi-

5

Research indicates many

areas in which general

education policies and

funding priorities are

having a negative impact

on maths and science

34

F R O M L A G G A R D T O W O R L D C L A S S

cial if the language involved is English, though learners with Afrikaans as a first language

seem to experience little difficulty in proceeding to grade 12, and managing any necessary

transition to English after that. How this issue is best approached and what steps ought to

be taken should be a priority task of a National Task Force (see recommendation 10).

The Dinaledi programme should be reconceptualised,restructured, and expanded.

Dinaledi has broken new ground for maths and science in the public schooling system.

However, in reviewing what has been achieved (and acknowledging the ‘virtues’), we

must also acknowledge where our performance has been disappointing, and take remedial

steps, so that many more schools and learners can participate and benefit.

CDE recommends that the Dinaledi programme be reconceptualised, restructured, and

expanded. It should be a permanent feature of the education system. It should also fall

under the aegis of the proposed National Task Force (NTF) on Maths and Science Educa-

tion (see recommendation 10 below).

Review all other educational policies for their effect onmaths and science.

As South Africa enters its second democratic decade, we are convinced that maths and

science schooling is its top educational priority. If one accepts this proposition, then this

has consequences for existing policies, approaches, and the allocation of resources.

Our research has indicated many areas in which general education policies and financ-

ing priorities are (unintentionally) having a negative impact on maths and science educa-

tion.

This situation should receive the urgent attention of our proposed National Task Force

(see later), in consultation with stakeholders. It should commission focused research, for-

mulate specific proposals for change, and submit these as soon as possible to the minister

of education.

The private sector and NGOs should review the support theyhave given to maths and science education with a view toaligning with the proposed national thrust.

Concerned private sector institutions should use this report as a basis for a discussion on

how to focus their input more strategically. Without being prescriptive, but in the spirit of

the systemic reform advocated here, the private sector should consider shifting its focus

from small-scale research and programme implementation to some or all of the following:

• Institutional support for a new public–private partnership to double the number of HG

passes in five years. In other words, help to fund the proposed National Task Force (see

later).

• Support to individual schools that are performing well and/or improving, or to schools

aspiring to join the specialised maths and science programme.

6

7

8

CDE proposes the

formation of a

public–private

partnership in the form of

a National Task Force to

drive the improvement of

maths and science

education in South Africa

35

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

• Support for an educator development programme based on credible international mod-

els.

• Support for programmes to identify learners with maths and science potential by

means of assessments in grade 9, ie, the aptitude test proposed earlier.

• Financial support for learners with potential who need to travel to or board at well-per-

forming maths and science schools.

• Provision of financial and other incentives to the best performing educators and learn-

ers.

• Financial support for maintaining and updating the CDE developed database on individ-

ual and school performance with respect to maths and science, as a tool for monitoring

progress. The database and its development should be placed under the control of the

National Task Force (see later).

• The work already being done by many corporations in providing bursaries for higher

education is acknowledged. Extension of these programmes to, or a tighter focus on

potential maths and science educators could be considered.

International aid agencies and foreign national donorsshould forge links with the new national initiative, anddevelop synergies between themselves and otherstakeholders.

Since 1994 financial contributions to South African education by international aid agen-

cies and foreign national donors have far exceeded those of local businesses and other

donors. The continued support of these foreign agencies will be essential for the success of

any maths and science programme. We believe this is a good moment for international aid

agencies and foreign national donors to commit additional resources and target their sup-

port behind the proposed national initiative and integrated programme of action being

suggested here.

The cabinet should establish a National Task Force as thevehicle to focus and direct a national partnership todramatically change the future of maths and scienceschooling in South Africa.

In making this recommendation, we aim to focus the commitment to adopt a systemic

nation-wide approach in a new institution devoted entirely to achieving the country’s

maths and science goals. Besides national and provincial government, there are a signifi-

cant number of stakeholders with an interest in this task. At present, they are not for-

mally involved in meeting this great challenge, except insofar as they are brought into

government initiatives or pursue their own (necessarily much smaller) initiatives. They all

do good work, but the whole effort is not greater (and may be smaller) than the sum of the

parts. Their efforts could be harmonised and amplified by creating a broadly based institu-

tion with responsibility for the whole system of maths, science, and language education.

Specifically, CDE proposes the formation of a public–private partnership in the form of a

National Task Force for the improvement of maths and science education in South Africa.

9

10

South Africa is dealing

with a national crisis. A

concerted new approach

by both public and

private sectors is required

36

F R O M L A G G A R D T O W O R L D C L A S S

Goals

The NTF should be made responsible for achieving two specific goals:

• doubling the number of school-leavers with an HG pass in maths, physical science, or

both within five years; and

• doubling the number of adequately qualified and trained educators in these subjects

with the same period.

These are the indispensable first building blocks of an incremental approach to systemic

reform.

Functions

The NTF should have specific functions that go well beyond ‘advising the minister’, and

generally promoting the need for an effective maths and science education system, though

both of these activities are necessary and desirable. It should also:

• Articulate and promote a strategy for achieving the goals referred to above, in a form

acceptable to the largest possible number of bodies or institutions presently working in

or providing funds for maths and science education. These should include all levels of

government; independent schools; higher education institutions; researchers and pro-

fessionals; donors; the South African private sector; NGOs active in education; maths,

science, and language educator associations; and others that make themselves known.

The aim is to achieve a greater alignment of effort than is presently the case, though

without discouraging initiative and experimentation with alternative approaches.

• Act as the body receiving and approving applications by schools to be appointed as par-

ticipating schools, and review the continued membership by each school after appro-

priate periods.

• Develop and administer a national assessment of learners’ mathematics and science poten-

tial, and link successful candidates to the nearest competent maths and science school.

• Develop and administer a national incentive programme for learners, educators, and

schools that will encourage them to participate positively in the national maths and sci-

ence programme.

• Maintain and update the school performance index created by CDE, and promote its use

in setting targets, designing methods, setting priorities, and assessing outcomes. These

potential uses are discussed in depth in the main report.

• Give substance to the urgent need to continuously monitor and evaluate the system.

There are five dimensions to this:

- Create and administer a credible assessment body that will give grade 9 learners a

chance to volunteer for an assessment of their maths and science aptitude. This will

enable parents, learners, and schools to make more informed decisions about choice

of subjects for SC.

- Commission ongoing and occasional monitoring and evaluation of aspects of the

system in consultation with the NdoE, provincial departments, and other bodies.

- Maintain a permanent monitoring and evaluation mechanisms for the system as a

whole, to assess progress and advise on incremental adjustments.

- Promote research into maths and science education trends, nationally and interna-

tionally, and bring findings relevant to the South African system to the attention of

the minister of education and/or other parties.

A bold response from

government is required.

If this happens, a

dramatic increase in

performance is

achievable in five years

37

- Create and maintain a programme for communicating with interested audiences

about the importance of maths and science education; the progress being made in

improving it; the logic of the strategy being followed; and stories of specific suc-

cesses. The programme should include an annual report to the cabinet, parliament,

and the public on activities and progress.

Membership

The NTF is envisaged as an expert (not representative) body, with public and private partici-

pation. The co-chairs should be drawn from the South African cabinet and business sectors.

Staff

CDE proposes the appointment of an executive director and one assistant only. All other

functions should be outsourced.

Funding

Initial funding should be sought from the public sector, the South African business sector,

and international and foreign donors for a five-year period.

CONCLUDING REMARKS

South Africa is dealing with a national crisis in its maths and science schooling. The cor-

rect response to this worrying situation is to adopt a concerted new approach by both the

public and private sectors. The key challenge is how to change a large and diverse system

without disrupting those parts of the system that are working.

CDE’s ten recommendations have far-reaching implications, especially as they would

play themselves out over time. They are practical, do not require extensive further

research, and can be implemented incrementally. We have set out to provide a broad

framework for action which is not a rigid blueprint to be followed and implemented

blindly. The proposed approach is a flexible one which takes account of existing public and

private interests and involvement in maths and science education. It also takes into con-

sideration the responsibilities and functions of different players from government at

national and provincial levels to parents, learners, educators and schools. In most cases a

start could be made at once.

Communication of this new approach, its benefits for all South Africans and its ambi-

tions within current severe constraints is an important consideration as we move forward.

The overall image of the new approach must be one of first building the higher level capac-

ity in sufficient quantity and quality to develop the essential expanded capacity which will

in turn allow a future transformation of the whole maths and science education system.

The CDE report provides a moment of opportunity for government. Key organisations

and interests in the private sector are willing and interested to help make a significant dif-

ference, hence their support for this privately funded initiative which has also received

enthusiastic support beyond its original donors.

A bold response from government is required. If this happens, a dramatic increase in

performance is achievable within a five-year period.

R E F O R M I N G M AT H S A N D S C I E N C E E D U C AT I O N I N S A S C H O O L S

South Africa has to deal with a national crisis with

respect to maths and science schooling. Current public

and private sector efforts are insufficient to significantly

change the system that is failing individuals, families,

communities, and the country

We must introduce a comprehensive programme that

provides for systemic change, involves public and

private sector leadership and resources, achieves short-

term results, and is implemented via an effective and

accountable institution

CDE proposes the formation of a public–private

partnership in the form of a National Task Force for the

improvement of maths and science education in

South Africa

This project has been funded by the Anglo American Chairman’s Fund, the

AngloGold Ashanti Fund, the BHP Billiton Development Trust, the De Beers

Education Trust, the FirstRand Foundation, the Joint Education Trust, the Liberty

Foundation, Murray & Roberts Holdings Limited, the Shuttleworth Foundation,

and the Zenex Foundation.

PREVIOUS TITLES

1. Post-apartheid population and income trends: a new analysis (September 1995)

2. South Africa's small towns: new strategies for growth and development (May 1996)

3. Cities and the global economy: new challenges for South Africa (October 1996)

4. Durban: South Africa's global competitor? (October 1996)

5. The East Rand: can South Africa's workshop be revived? (June 1997)

6. People on the move: lessons from international migration policies (June 1997)

7. People on the move: a new approach to cross-border migration in South Africa (June 1997)

8. Pretoria: from apartheid's model city to an African rising star? (July 1998)

9. South Africa's 'discarded people': survival, adaptation, and current policy challenges (October 1998)

10. Policy-making in a new democracy: South Africa’s challenge for the 21st century (August 1999)

11. Johannesburg, Africa’s world city: a challenge to action (October 2002)

12. Key to growth: supporting South Africa’s emerging entrepreneurs (June 2004)

Cover photograph by Gisèle Wulfsohn/South Photographs

Produced by Riaan de Villiers and Associates

CDE RESEARCHPOLICY IN THE MAKING

BOARD

E Bradley (chair), F Bam (deputy chair), S Ndukwana (deputy chair),

A Bernstein (executive director), N Angel, F Antonie,

C Coovadia, O Dhlomo, W Esterhuyse, M Keeton, A Lamprecht,

J Latakgomo, R Lee, P Lourens, A Mandewo, J McCarthy, R Menell, I Mkhabela,

K Mthembu, M Mthembu, W Nkuhlu, A Oberholzer, M O’Dowd, F Phaswana,

R Plumbridge, D Ramaphosa, L Schlemmer, N Segal, J Seutloadi,

C Simkins, G Smith, M Spicer, T van Kralingen

INTERNATIONAL ASSOCIATE

Peter L Berger