METRO CENTRAL EDUCATION DISTRICT
LEARNER RETENTION IN THE SOUTH AFRICAN SCHOOLING SYSTEM
A MINISTERIAL REPORT
In considering the provisions of the legislative framework, it is evident that the Department has a responsibility to account for all children of school-going age, whether they attend school or not. The investigation into learner retention in the schooling system has explored a number of indicators of the performance of an education system in retaining learners, including survival-rate, age-specific enrolment rate, general levels of education attainment in the population, as well as the frequently cited dropout and repetition rates. These measures highlight the efficiency of an education system from different angles, and the use of one without the other, does not provide a comprehensive overview of the education system. Data sources used include EMIS, Census, Labour Force Survey and General Households Survey in an attempt to obtain a holistic view of the extent of learner retention. The Community Survey 2007 was released at a stage when the Committee had finalised its analytical work. In the process of the investigation, the Committee employed a multi-faceted approach that includes the following:
? Quantitative analysis of population and schools data
? Empirical research review focusing mainly on empirical studies that were published in recognised research journals or were conducted by scholars in the education sector
? Methodology assessment, to address point 1 in the Terms of Reference, which included discussions with researchers, commentators and journalists who have conducted studies on the subject of learner retention and learner dropout, or have written public commentary on the subject; and
? Public submissions in order to obtain an understanding of the reasons behind the phenomenon of leaving school before significant milestones are reached. It is important to highlight that the submissions provide useful information, which cannot, however, be treated as conclusive evidence due to the size of the sample.
With regard to the extent of retention and dropout, having reviewed existing data, information and research, the investigation concluded that there is sufficient evidence to state that:
There is a problem of learner retention, which is more pronounced after Grade 9. The dropout rate below Grade 9 is statistically insignificant, but increases sharply from Grades 10 to 12. A proportion of learners starting Grade 9 are not in a position to finish secondary school, and that the system does not provide sufficient alternatives. As a consequence, there is a high failure rate, repetition and dropout in Grades 10 – 12, which is a waste of many years of learning. Urgent attention should be given to providing suitable alternative FET programmes which include a focus on both content and mode of delivery. There is no evidence of anomalies between Grade 1 and Grade 2 that point toward dropping-out. In fact, as can be seen in the survival rates by grade, the flow between the two grades is just below a hundred percent. A problem that seems to plague Grade 1 is high repetition of the grade, a phenomenon which is not peculiar to South Africa. However, retention is improving, as are levels of education attainment, with younger age groups having a better chance of progressing to secondary school grades than the older age groups. South Africa compares favourably with other developing countries on progression rates, enrolment rates and levels of education attainment.
Learner retention at FET level indicates that the percentage of people with Grade 9 reaching Grade 12 tends to remain almost static over the years. The estimates indicate a static progression pattern within further education among learners entering this phase with slightly less than 90% of those with Grade 9 reaching Grade 10. About three-quarters reach Grade 11 and just below 60% reach Grade 12. Essentially, the drop-out rate is minimal for at least the first 8 years of schooling. The dropout rates increase sharply from Grade 9 onwards. Out of a thousand people born between 1980 and 1984, about 984 entered Grade 1 and 456 reached Grade 12. This means slightly more than 46% of this birth cohort that started Grade 1 eventually reached Grade 12.
Mean educational attainment of the cohort born in 1981 is about 9½ years, compared to only about 6 years of the cohort born in 1945. (As the latest data is drawn from the 2001 census, it is not possible to investigate the situation for cohorts born after 1981, as many of these would not yet have completed their education by 2001). As a result of rising educational attainment, the proportion of the population reaching certain educational milestones has risen dramatically. For example, the cohort born in 1981 is approaching universal completion of primary education, as has long been the case for Whites. When the lowest schooling hurdle – completed primary education – is used, the racial gap in attainment is quite small. There is only a difference of 14% between the proportion of Whites and Black Africans who have completed primary school. (Higher milestones, such as matric and tertiary qualifications, show a dramatically different pattern, though Matriculation completion rates are low amongst the Black African and Coloured population, even though these have been rising quite significantly over the years). There is an upward shift in the entire educational attainment profile for each of the censuses for the 21 to 25 age group. This upward shift is greatest in the case of the Black African population. From the 21 to 25 year-old Black African people in the 1970 census, 40% did not enrol for school and fewer than 1% passed matric. By the 2001 census, these figures had improved to 9% and 36% respectively for the same age group. The attainment profiles of Black African children aged 16 to 20, according to their parents’ income quintile, rises strongly with higher per capita income. Age-specific enrolment figures reveal a fairly consistent pattern - sustained high enrolment rates of 95% or above until about age 15 or 16, following which enrolment falls quite sharply as expected to about 50% by age 19. This pattern is so consistent that there can be little doubt that there is almost universal school enrolment until age 16, which corroborates the findings from the analysis of survival rates and dropout rates. There is no clear evidence of changes in age-specific enrolment patterns over the past decade. Age-specific enrolments also show no statistically significant evidence of dropout from the school system before age 16. Age-specific enrolment figures at age 15 or 16, broken down by race, indicate that the Coloured population shows the earliest trend towards drop-out, before the completion of high school. The Black African population evinces a major dropout rate at about age 16. The White population has the lowest dropout rate. No separate figure is shown for the Indian population, as their small dimension makes it impossible to use sample surveys of this nature as good evidence of enrolment patterns. While the data indicates that there is minimal dropping-out of the school system before the age of 16, (at least not for about 95% of the population), there are clear indications that learners show high levels of perseverance at secondary schools, without achieving matriculation, when the age restriction comes into effect. In addition, the pattern of age-specific enrolment indicates that, for the most part, all race groups have been attending school for the past decade until well beyond the compulsory phase.
Learner Retention in the South African Schooling System is available to provide an international frame of reference. Patterns of progression through the South African system are favourable compared to this group of middle-income countries with which SA is often compared. Venezuela, the Dominican Republic, Panama, Brazil and Nicaragua have shown lower education attainment levels for this age group than South Africa. Bolivia and Peru have shown generally higher attainment levels than South Africa (Data used pertains to different years due to data availability). With regard to the reasons for dropping-out among learners, the investigation found that: Grade repetition has been identified as the single most powerful predictor of dropping-out. Studies conducted internationally have revealed that learners who have repeated a grade in their schooling career are most likely to drop out of school. Learners become disillusioned, and generally disengage from school activities. The fact that grade repeaters are taken through exactly the same material and content when repeating the grade, exacerbates the situation.
Grade repetition occurs most frequently in the first grade. Schools all over the world experience higher grade repetitions at the start of a school cycle than they do in subsequent years. The high rate of grade repetition in the first grade is attributable to inadequate school readiness programmes, serious problems with learners’ learning abilities, or significantly high enrolments which have not been accompanied by appropriate levels of provisioning. Drop out is preceded by indicators of withdrawal or unsuccessful school experiences (academic or behavioural difficulties) which often begin in the primary school years. Preventive measures should be taken at the earliest manifestation of withdrawal indicators. The risk of dropping-out is very high for learners who are older than the median age when they enter secondary education; independent of their achievement scores (therefore grade repetition needs to be minimized). The cycle of previous failure may either impact on these students or they may have more mature goals than their younger classmates, goals which are non-school related, such as obtaining an occupation, an income, raising a family, or goals that are more directed at autonomy. Educational levels of parents related strongly to dropout and interacted with gender. Girls with highly educated parents have a lower risk of dropping-out than boys with highly educated parents. The effect of educational level also differs for minority and non-minority groups.
All the methodologies that were widely quoted in the media in the recent past were flawed. They ranged from crude calculations which simply subtracted the number of learners in a grade from the number of learners in the next grade of the following year, and declared the difference in dropouts - to formulae that used Grade 12 enrolment as a proportion of Grade 1 enrolment twelve years ago. Such crude calculations have omitted critical variables, including: Policy provision which allows learners to repeat grades for a maximum of four times in their school career. The Admission Policy for Public Ordinary Schools requires the norm of repetition at one per school phase, effectively allowing learners to complete a 12-year schooling career in 16 years.The progression of a significant number of Grade 9 learners into public Further Education and Training (FET) colleges to enrol for Grade 10 – 12 equivalents. Other forms of learning institutions such as private FET colleges, home schools and workplace-based colleges where data is not easily available to detect the extent of coverage of the 15 to 18 year old group. A more rigorous method is the cohort model presently used by the Department of Education, which is also unsatisfactory both in respect of its structure and in respect of the estimation of repeater and dropout rates.It arrives at dropout rates by subtracting the promotion rate, plus repeater rate, from one. This makes the dropout rate a residual whose value is sensitive to errors in the promotion and the repeat rate. Since repeater information is almost always underreported, the low repeater data will invariably lead to exaggerated dropout rates. In addition, the use of a school-based EMIS system to trace children who are out of school creates additional hurdles. A learner-based tracking system would yield more reliable findings. A number of technical challenges were encountered in using the EMIS data to arrive at conclusive evidence on learner retention. The challenges related to the non-differentiation between a nil return in terms of the number of repeaters and “zero” number of learners who are repeating. In addition, the EMIS data does not constitute a complete accounting system and is thus flawed. Estimates based on it are subject to unknown selection biases.
Departmental estimates of survival based on aggregate repeaters taken from EMIS, are sharply biased downwards. These challenges would have been experienced by the Department in its calculations. The Committee concluded that the state of the Annual Schools Survey in EMIS makes estimates of repeater rates quite uncertain and estimates of drop-out virtually impossible.
So what are the alternatives?
In the face of the paucity of sound methodologies and education data, the investigation has relied on estimates of the population extracted from the General Household Surveys, the Labour Force Survey and the Census data. Estimates based on the General Household Survey are more reliable, but they are necessarily cohort based and therefore cannot be calculated with precision until most of the cohort has passed beyond school education.
In measuring the extent to which the schooling system is able to retain learners, the investigation used: the survival curve that measures the survival of a cohort from Grade 1 to Grade 12 or of any group of learners; extent of coverage measured by the age specific enrolment ratio; the levels of education attainment in the population in general; and grade attainment per age group to determine the retention rate over the past few years. These measures are used internationally to measure the performance of an education system, and the Committee strongly recommends that greater attention be given to these measures and less on dropout. In order to move towards improved measures of system efficiency, consideration should be given to new survey instruments that will enable South Africa to put the estimation of learner retention on a sounder empirical basis. If the intention is to measure internal efficiency, then consideration should be given to the widely used measures such as repetition rates, school life expectancy and years of input per graduate.
1.3 Key recommendations
On issues relating to the findings on the extent of learner retention and drop out in the schooling system, it is recommended that: Interventions to improve learner retention in the schooling system should focus on the post-compulsory school phase as there is conclusive evidence to suggest higher dropout rates from age 16 onwards. While it is noted that the age group is beyond the compulsory school phase, it is in the interest of the country to retain as many learners as possible until they complete Grade 12, or until they have completed an equivalent qualification through the FET College system. A small proportion of learners starting Grade 9 do not stand a good chance of completing Grade 12. The education system does not provide sufficient alternatives for the 16-18 year old group. Consequently Grade 10 – 12 have high rates of failure, repetition and dropout. Urgent attention should be given to providing suitable alternative FET programmes which include a focus on both content and mode of delivery. Despite the insignificant dropout rates in primary school grades, it is imperative that the Department investigates circumstances for learner absence as provided for in the South African Schools Act No. 84 of 1996. Active mobilisation of communities to identify and report children who do not attend school should be encouraged. With regard to reasons that lead to learner drop-out and the interventions that have to be designed to prevent learner drop-out, it is recommended that: Consideration be given to effectively utilizing the wider communities and improving social networks to monitor and track learner attendance. In addition, consideration should be given to developing a cadre of “attendance officers” who should be appointed at local levels to monitor attendance and provide psychosocial support to learners who are at risk of dropping out.
Indicators of low self-esteem and aggressive behaviour in early childhood should be attended to so as to promote continuing academic success and prevent dropout. Early measurable factors and behaviours are highly associated with later school drop-out. Grade repetition is generally ineffective as an intervention to address early learning problems, regardless of when the repetition occurs. Learners repeating grades should have special programmes that are not a mere repetition of the material and content covered during the first year in the grade. In this regard, lessons can be learned from the Grade 12 recovery plans which were implemented in schools across the country following the month-long public service strike that took place in May/June 2007, provided that the programmes are well researched and understood. Access to early childhood development be improved. There is a growing research/knowledge base that demonstrates that children who have experienced ECD interventions, or minimum pre-primary schooling, are better achievers at school than those who have not experienced ECD interventions or minimum pre-primary schooling. Those children who attend ECD programmes are more highly motivated, perform better, achieve higher scores in cognitive tests, and socialize better with their classmates and teachers. ECD graduates are therefore less likely to drop out or to repeat grades. Therefore, the cost of their schooling is reduced, with the result that primary and even secondary education is more cost-effective. Thus, ECD in itself can spur educational participation in a region of the world that lags behind in most educational indicators. (Hyde, 2006). South Africa is committed to ensuring that by 2010, all learners entering Grade 1 would have completed an accredited Grade R programme. It is therefore recommended that the roll out of Grade R be prioritised and that the necessary ground work be completed without delay. Positive classroom climate should be actively created and schools should be urged to cultivate supportive personal relationships with struggling students. Smaller class sizes, more personalised settings and individualised learning plans are identified as characteristics for lowered drop-out rates in some of the studies. It is recommended that a more intensive interventional approach should be applied. Intensive staff development programmes for teachers should be provided in order to improve their skills and ability to identify learners with learning disabilities. Learners who are at risk of dropping out of school should be identified and the use of a variety of instructional and assessment methods and techniques should be applied in order to benefit the largest possible number of learners in class.
On issues of data quality, as well as the improvement of data collection and reporting systems, it is recommended that: Particular attention is paid to improving the data collection systems of the Department of Education. Interventions for improving the systems should begin at school level, ensuring that all schools submit all required data every year. Consideration should be given to linking submission of forms to resource allocation and making the system an accounting system so that schools that did not submit the previous year are restricted from submitting