/ 11 January 2019

Pass rate says little about schooling

Anxious business: Grade 12 pupils check to see if they have passed matric. Photo: Fredrik Lerneryd
The results of learners implicated in the leak of the two papers in the 2020 matric examination and the results of those learners involved in “group copying” will be blocked while investigations continue.

The frenzied debates in January each year that follow the release of the matric results attest to the nation’s passion for education and our cherished freedom of expression. But this passion should be combined with more analytical rigour.

Every year there is a squabble over what the “true pass rate” is, meaning National Senior Certificates (NSCs) obtained by an age cohort of the youth population. The fact checkers at AfricaCheck conclude that in recent years just over 50% of youths obtained the NSC. This is credible.

What is not credible, but common, is dividing NSCs by the enrolment in a lower grade in an earlier year. This could work if no pupils repeated grades. But with our high and varying levels of repetition, this method renders nonsensical statistics.

The most recent culprit is the Western Cape education department. Using this bogus method, they concluded, incorrectly, that their “real” pass rate in 2018 was the highest in the country. In fact, black (meaning coloured and black African) youths in the province still have a lower probability of obtaining the NSC than black youths in Gauteng, Mpumalanga and KwaZulu-Natal, though the Western Cape has been catching up.

But, even if it is just below 50% of youths who fail to obtain the NSC, and not 63% as suggested by the Democratic Alliance a year ago, this is still a staggering figure. What should be done?

The arguments imply that we should realise universal completion of 12 years of schooling soon, with the proviso sometimes added that an increasing proportion of this schooling should have a strong technical or vocational orientation. This sounds reasonable, but we should ask how fast it typically takes developing countries to achieve this. In fact, the process is slow. Moreover, our current secondary school completion ratio, of just over 50%, is about average for a middle-income country. Between 1994 and 2018, we pushed this statistic up by 0.8 percentage points a year, exactly the speed of progress seen in relatively successful Malaysia during the 30 years between 1970 and 2000.

Why do youths drop out before grade 12? If the main reasons were financial, the problem would be relatively easy to solve. But the 2017 General Household Survey indicates that 64% of grades eight to 12 pupils pay no fees because of the government’s pro-poor no-fee policy. Moreover, 78% of secondary school pupils receive lunches as part of a school feeding scheme. So, the factors are not mainly economic.

The foremost reason for dropping out is pupils not coping with their schoolwork. The remedy for this is better learning and correct teaching from the initial grades, something that has happened, according to the international testing programmes. The complex task of continuing with this qualitative improvement is what is needed to increase the number of NSCs further.

But if achieving universal upper secondary school completion is a gradual process, what should be done about the dropouts in the interim? Botswana and Namibia provide some lessons. In these countries, national examinations and a qualification exist in grade 10, providing those who do not achieve a grade 12 qualification with something to fall back on to rate their competencies for employers and post-school training institutions.

Namibia’s Junior Secondary School Certificate was introduced a decade after independence. Our 1995 education white paper envisaged a similar grade nine General Education Certificate (GEC). The 2014 report of the ministerial committee looking into the NSC reiterated a need for a lower secondary qualification. The fact that this is barely mentioned in the national debates is striking.

One thing a GEC would facilitate is the transition of pupils from schools to technical and vocational education and training (TVET) colleges. Currently, this transition is inefficient. By far most of what is offered in the colleges is at the level of grades 10 to 12 in schools, in terms of the National Qualifications Framework. The policy implies youths should transition from grade nine in a school into TVET in a college.

But, because grade nine pupils are unable to demonstrate their competencies easily for lack of a national qualification, colleges veer towards the safer option of admitting primarily grade 12 graduates with the NSC. This largely explains why the average age in TVET colleges is 24, rather than 18 in grades 10 to 12 in schools.

When the official NSC “pass rate” — NSCs divided by candidates — is announced each year, there are immediately concerns about whether this can be relied on as an indicator of progress. Is it being manipulated for political ends, many ask?

These concerns miss the point that exams are poor gauges of systemic progress, for instance because subject choices and participation rates shift from year to year. The primary aim of an examination is to provide people with qualifications. There are good reasons why the matric pass rate is not among the set of education indicators in the government’s Medium-Term Strategic Framework.

What ought to receive more attention are South Africa’s results emerging from international testing programmes such as the Trends in International Mathematics and Science Study. These are specifically designed to gauge progress in the system and have pointed to considerable improvements during at least the past 10 years, though they also suggest progress has slowed down in recent years, years that have seen exceptional budget constraints in public schooling.

Some of the energy spent worrying about matric results could be spent more productively by looking at what kind of national sample-based assessment system we should have to complement the international programmes. Current efforts to set up the National Integrated Assessment Framework represent what is arguably the most important basic education policy initiative under way but it has received scant attention in the public debates.

Outrage is often expressed about the fact that the lowest possible pass mark per subject is 30% (though what is not clarified is that at least 40% must be achieved in three of seven subjects).

The lowest threshold has been 30% since 2008 and before that it was 33% for several decades. This is thus something deeply entrenched in the schooling system.

Arguments are made that raising this threshold would improve the education system. The 2014 ministerial committee, which recommended several changes to the NSC, many of which have been followed through, did not recommend changing the lowest threshold. There are good reasons for this. Perhaps most importantly, disruptions would result in one youth cohort experiencing different levels of access to the NSC than another, which would be unjust.

Universities and employers are less worried about the level at which thresholds are set than that subject-specific marks should be stable over time. A mark of 60% in mathematics in 2011 should mean the same as a mark of 60% in 2018. This is what allows, for instance, university engineering faculties to admit only those people able to cope with the programme. On the whole, this type of stability has been rigorously maintained.

Provinces are increasingly holding secondary schools accountable against measures such as the number of pupils achieving 50% in mathematics. How to get such strategies right deserves wide public attention. And such strategies do not require disruptive changes to the structure of the NSC qualification.

Martin Gustafsson is an associate professor at the University of Stellenbosch, where he is a member of the Research on Socioeconomic Policy group. The opinions expressed here are his own