Last month the national benchmark tests initiative, designed to address the alarming failure rates in higher education, found it necessary to reset its standards. This move reveals increasing concerns about the educational needs of students as they enter higher education and a growing acceptance that educational disadvantage is a common phenomenon in the system.
Higher Education South Africa, the vice-chancellors' representative body, commissioned the national benchmark tests project in 2005 to provide information about the levels of knowledge in core areas of incoming students.
It is generally acknowledged that South Africa's education system is in trouble. Historically, the lightning rod for this has been matric, which until recently has been the only externally moderated national examination in the South African education system.
However, there are indications of serious problems at all levels of schooling and higher education. In schooling, the newly introduced annual national assessments for grades one to six — and from this year also grade nine — are showing the pervasive and extensive problems facing our schools. For example, the basic education department's report on last year's assessments indicates that less than a third of all grade six learners perform at a level that indicates even partial achievement of curriculum competencies.
This dismal level of performance continues into higher education, where about 40% of registered students drop out of their studies during or at the end of their first year and only about 15% obtain their degrees in the minimum time.
This clearly unacceptable situation has led to a national focus on "throughput rates" and to greater acceptance of four-year degrees, a possibility that is now being investigated by the Council on Higher Education.
Although more realistic curriculum structures such as a four-year degree would undoubtedly represent a step forward, the so-called articulation gap between schooling and higher education continues to pose a challenge. This is so particularly in the context of continuing curriculum changes in grades 10 to 12 — changes that are not adequately supported in terms of teacher preparation or learning materials, as the recent textbook debacle in Limpopo has demonstrated.
The national benchmark tests attempt to clearly measure the underlying knowledge, skills and cognitive abilities that students will need for higher education and, by so doing, to provide feedback in two directions: to schooling, about the kinds of preparation that are needed, and to higher education, regarding the curricula that are likely to be appropriate.
In this undertaking the outcomes of the matric exam in key areas are taken as the departure points while at the same time a clear focus is maintained on higher education's expectations and assumptions about how this knowledge and ability will be used.
The core question posed in the development of the national benchmark tests is: To what extent do school leavers aiming to enter higher education meet the core academic literacy, quantitative literacy and mathematics competencies required to study at tertiary level?
The tests do this by providing a score as well as a proficiency level in each of these three areas. The process uses academics from across South Africa to develop benchmarks that place a student into one of three performance categories for each test domain:
• Proficient: Performance in the domain area suggests that the student's academic performance will not be adversely affected. If admitted, the student should be placed on regular programmes of study.
• Intermediate: Challenges in domain areas are identified, indicating that a student's academic progress will be affected. If admitted, the student's educational needs must be met in a way deemed appropriate by the institution, for example, by introducing extended or augmented programmes and providing special skills.
• Basic: Serious learning challenges are identified that indicate a student will not cope with degree-level study without extensive and long-term support, perhaps best provided through bridging programmes or further education and training colleges. Institutions registering students performing at this level would need to provide such support.
In 2009 the project made headlines when its results showed the proportions of students needing some kind of intervention. The size of the challenge was made more significant because the new matric exam — the national senior certificate, written for the first time in 2008 — showed very different results, particularly in mathematics.
Students' performance in the first year of degree study from 2009 to 2011, however, suggests that the national benchmark tests' degree benchmarks that were set in 2009 were effective in identifying those students who would be able to complete a degree in the minimum time, those who would need additional support and those who would not be able to cope with degree study even with additional support.
Changes in degree benchmarks
Last month's setting of standards involved two exercises. The first one reset the degree benchmarks and the second determined pilot benchmarks for diploma study. The most striking changes in the degree benchmarks are that:
• For all three domains (mathematics, quantitative literacy and academic literacy), the proportion of pupils who require support has grown; and
• In mathematics, the score dividing "proficient" from "intermediate" has moved upwards from 62% to 68%.
The second finding represents a quite dramatic increase and indicates the strong concerns and challenges that are experienced in the system relating to the preparation level of students entering mathematical disciplines. This concern suggests that extended mathematics offerings, such as doing first-year mathematics over two years, should be the norm rather than the exception.
The impact on the proportion of test writers meeting this new benchmark is significant: the previous benchmark was met or exceeded by about 8.5% of all national benchmark test maths writers, whereas with the new benchmark the proportion will drop to 5%.
In other words, academics estimate that only 5% of students can be described as proficient and needing no additional support.
Two things are clear: the schooling system still does not adequately prepare students for the demands of higher education study and curriculum structures are not able to address the needs of the majority of students wishing to embark on tertiary study.
Robert Prince is director of the alternative admissions research project and Professor Nan Yeld is dean of higher education development at the University of Cape Town