Schooling , the underclass and intergenerational mobility : a dual education system dilemma

School education in South Africa has seen much progressive change in the last 20 years. Yet educational outcomes are poor and many argue that a dual education system exists. Those with financial and socio-cultural capital access resourced schools, while poor South Africans are relegated to schools still suffering from apartheid resource neglect. This empirical study of high schools in Alexandra township, a poor black African residential area, demonstrates both the extent of the resource backlog and the consequences thereof. Secondary schools in Alexandra have an inadequate number, and standard, of toilets, libraries, computer facilities and science laboratories. They also have relatively high learner to teacher ratios and poor matriculation success rates. Enrolment in such schools means learners achieve a poor quality matriculation certificate or none at all, thus, trapping these learners into significant disadvantage. Meagre financial resources preclude Alexandra parents from selecting better resourced schools. Thus, for these learners, neither their legal rights with respect to school choice nor their geographical proximity to resourced schools has ensured redress from the apartheid past. The result is that intergenerational class mobility is limited. Thus, the dual nature of South Africa’s education system is creating a vicious cycle of intergenerational poverty where young people cannot improve their living standards despite enrolment in secondary schooling.


Introduction
It is often argued that South Africa is one of the most unequal nations on earth (Leibbrandt et al, 2010).Without doubt, former racial discriminatory practices are largely to blame, although others argue that specific pre-and post-1994 state policies have exacerbated inequality by increasing the rate of unemployment (Nattrass & Seekings 2001, Crankshaw, 2008;Bond 2011).The human cost of this inequality is borne most heavily by poor black African a people, who, due to apartheid era spatial discrimination, live in geographically marginalised communities.These individuals often eek out a living in the informal sector or survive on social welfare transfers.For those who are formally employed, they are often working-but-poor, partly because they lack the skills and qualifications to access well-paid positions (Crankshaw, 2008).Rightly or wrongly, then, South Africans pin their hopes for a better life on education (Maile 2004).In particular, access to resourced schools is viewed as a means to rectify apartheid wrongs, a human right and a panacea for unemployment.That is, enrolment in a resourced school is assumed to provide an individual with the skills and knowledge required to access the world of formal work, to enable a person to demand higher wages and enable access to tertiary education (Bhorat 2004).
As education is viewed as crucial to intergenerational class mobility, the post apartheid government focussed much time and energy on educational reform.Thus, many schools are now racially desegregated.There has been significant redress in how public funds are allocated to schools.Parents, via school governing bodies, can now provide greater inputs into how schools are managed.The provisions of the South African Schools Act, Act 84 of 1996 (SASA) allow learners, regardless of social status and geographical location, to access any public school they like.Practically, however, such a wide ranging right is difficult to administer.Thus, school admissions are regulated through the provisions of the National Education Policy Act, Act 27 of 1996 (NEPA).NEPA regulates school admissions using a combination of geographical catchment zoning and designated feeder schools.It also allows school governing bodies to levy school fees and determine the language of teaching and learning (Bell & McKay 2011).Problematically, due to past apartheid educational policies, resourced schools are in critical short supply, with thousands of schools formerly allocated to black, coloured and Indian people, lacking financial, infrastructural and human resources (Fataar 2008).The result is that demand for resourced schools far outstrips supply and NEPA is often used to limit enrolment numbers in Gauteng (Bell & McKay 2011).In such a situation, despite SASA and the school fee waiver system, those with financial and sociocultural resources manage to enrol their children in resourced schools, whilst poor people are being excluded (Weber 2002;Lemon 2004).As a result, scholars have argued that poverty will persist, as structural exclusion from educational resources entrenches intergenerational inequality (see Kallaway 1984, 1997, Nattrass & Seekings 2001, Sekete et al 2001, Seekings & Nattrass 2002, Louw 2004, Lipton 1989, Maile 2004, Fleisch 2008).
The last 20 years has seen South Africa make significant strides in terms of changing school education (Soudien 2007).For example, South Africa now boasts almost universal enrolment, quite a feat considering the millions out of school prior to 1994 and rapid population growth (Fataar 1997;Gustafsson & Patel 2006).Nineteen separate education systems have been collapsed into one and the school curriculum has changed (numerous times) (Bloch 2010).All of which have gone a long way to rectify apartheid wrongs.However, apartheid education also extended to the grossly unequal provision of teachers (in terms of both numbers and qualifications), physical resources and school management capacity (Nattrass & Seekings 2001;Fataar 2008).Transformation in terms of teachers, physical resources and school management systems, is proving to be much more challenging.This is partly due to funding decisions.Despite the massive increase in money spent on each black learner post-1994, overall expenditure per learner has not reached the same levels that the apartheid government used to spend on each white learner (Gustafsson & Patel 2006;Evoh & Mafu 2007).Many scholars argue that this is because of the adoption of neo-liberal economic policies (i.e.fiscal restraint) hampered full financial redress (Fataar 1997;Louw 2004;Soudien 2007;Hall & Giese 2008/9).In fact, between the years 1995 and 1998, the education budget declined in real terms (Ndimande, 2006;Fataar, 2008).This situation was later partly rectified and South Africa now spends 5% of its GDP on education.Still, funding norms have focused on equalising current school financing and not in addressing apartheid backlogs (Louw 2004;Maile 2004;Fiske & Ladd 2004;2006;Ndimande 2006;Redpath 2006). For Fataar (1997) and Lemon (2004), this emphasis on frugality meant South Africa has focused on quantitative not quality education expansion.Chipkin (2003:35) argues that such funding decisions resulted in the working class being "forsaken...in favour of the state and a black bourgeoisie".Clearly, then policy decisions relating to school funding norms need serious interrogation (Nattrass & Seeking 2001;Bond 2003Bond , 2004;;Louw 2004;Soudien 2007;Fataar 2008 andFleisch 2008).The lack of resourced schools means that learners who attend them do not acquire the knowledge, skills and qualifications required to access the world of work or tertiary education (Hunter 2010).Thus, such individuals become structurally excluded from the mainstream economy and poverty is entrenched over time within particular families.This is similar to the way that apartheid did, only this time it is not African people in general who are affected, but poor African people living in geographically marginalised or ghetto, communities, like Alexandra township, who are.Internationally, such a group of people are viewed as an 'underclass'.
Worldwide, poor residential areas have been perceived as a space where an underclass, that is, a group trapped in intergenerational poverty, either resides or is created (Wilson 1987).It is argued that such areas trap impoverished people, locking families into their socio-economic status by isolating them from mainstream society.However, the concept is both complex and controversial (Jargowsky & Yang 2006).Myrdal originally coined it in 1962 to explain the chronic structural poverty experienced by people in the United States of America (USA).For Myrdal, such people were, by virtue of de-industrialisation, rendered unemployed, underemployed or unemployable.This phenomenon of unemployment was structural due to a skills mismatch.That is between what skills such people had (if any) and what the service economy demanded (Lee 1994;Woodward 1995).Thus, the underclass are people who are victims of the so-called cycle of creative destruction inherent in the capitalist economic system.It is hard to reintegrate such people into the formal economy, so they become increasingly 'set apart' from the rest of society, eking out an existence on very low incomes (Gans 1993).For Miles (1989), Wilson (1987;1989;1996), Lee (1994) and Woodward (1995) the use of the term underclass signifies the systematic marginalisation of poor people because of state policies, society in general and capitalism in particular.This marginalisation becomes intergenerational if the children of such people are also structurally excluded from educational resources.The result is inequality, geographical marginalisation and social polarisation (Gephart & Pearson 1988;Hochschild 1989;Wacquant & Wilson 1989;Phillips & Karn 1991;Thornley 1992;Beauregard 1993 andDanziger &Gottschalk 1995).
The notion of an underclass gained widespread usage in the 1980s.Unfortunately, American journalist, Ken Auletta, 'hijacked' the term to use it to claim that individuals in the underclass become -and remain -poor because of socially deviant behaviours, not because of structural issues (Auletta 1982;McLanahan & Garfinkel 1989).As the majority of these underclass individuals in the USA were black, the phrase became loaded with racist connotations (Jones 1987).Politically conservative politicians began to use the term to describe and explain poverty-ridden groups of people.The result was that scholars such as Pahl (1988), Miles (1989) and Woodward (1995) called for the scrapping of the term -none more so than sociologist, Herbert Gans, who felt that the term misrepresented the ghetto poor (Venkatesh & Rosen 2007).It is largely due to this warping of the concept that the term has fallen out of favour worldwide (Jargowsky & Yang 2006;Venkatesh & Rosen 2007).Despite international vilification, the concept of underclass is used in South Africa (and elsewhere) by academics such as Crankshaw (1997), Bond (2003Bond ( , 2004)), Seekings andNattrass (2002), andLouw (2004).For these academics, the notion of a structurally generated underclass can help explain how poverty can become intergenerational.Their argument is that the post-1994 political dispensation did not put sufficient policies (one of which is education) in place to alter the structural gap between rich and poor, white and black.Thus, learners are "underserved and structurally positioned" by South Africa's dual education system (Louw 2004;Deil-Amen & DeLuca 2010:27).So, the revival of the underclass debate, internationally and in South Africa, serves to flag persistent poverty as a structural phenomenon or a type of 'class apartheid' (Seekings & Nattrass 2002;Bond 2004:3;Kamete 2006;Gove 2011;Newburn 2011;Cameron et al 2012).If poverty in South Africa is structural in nature it will require concerted state action to combat it, in the same manner racial apartheid was dismantled.This includes progressive policies specifically designed to help poor people residing in economically marginal areas such as Alexandria improve their living standards and their access to the job market.Without such interventions, the underclass will simply grow in size, becoming a significant destabilising force, which could undermine much of the post apartheid gains.

Methodology
Alexandra is a poor black dormitory township located in the northern suburbs of Johannesburg.Its low socio-economic status and lack of racial desegregation is in complete contrast to its geographical neighbour, Sandton, which a wealthy, multiracial elite call home (Crankshaw, 2008) (see Figure 1).

Figure1: Location of Alexandra township within the City of Johannesburg
Geographically, Alexandra is separated from Sandton by 'classic' apartheid spatial barriers such as main roads, highways and industrial areas (Lipton 1989).As it was originally located far from the Johannesburg CBD, it was shunned by white buyers and so developed (in 1912) into a 'native' township.Its racial character was preserved under the 1923 Native Urban Areas Act, which zoned it for people of colour.This made Alexandra unique, for South Africa, being one of the few areas were people of colour could legally live within 'designated white urban space' (Bonner & Nieftagordien 2008).Despite efforts by the apartheid government to forcibly remove them, black people continued to make Alexandra their home throughout the apartheid era (Mabin 1991;Parnell & Pirie 1991;Wilson 2002;Bonner & Nieftagordien 2008).In the history of urban South Africa then, Alexandra is legendary as it represents a victory against urban apartheid planning and enabled some black people to retain a toehold in valuable urban land.Currently it is estimated that Alexandra has a population of between 180 000 and 350 000 people, who reside in informal settlements, typical 'match-box' township houses, hostels and RDP-type units.In the 1980s, a new, 'middle-class' zone, namely the Far East Bank was developed (Bonner & Nieftagordien 2008).Generally, residents are poor, with an average monthly income of R1 029, and unemployment is rife (Wilson 2000).In many ways, Alexandra could be viewed geographically as a 'reservoir of poverty', or a ghetto, mirroring patterns in other countries, such as the USA and the UK (Pahl 1988;Mingione 1993;Lee 1994;Yapa 1996 &;Fainstein 2001;Crankshaw 2008).
This study sought to answer the following research questions: (1) To what extent are Alexandra learners trapped into choosing poorly resourced schools?(2) Do the schools in Alexandra perpetuate socioeconomic status by inhibiting the ability of their learners to enter the world of work and/or tertiary education?There are 18 schools in Alexandra, with this study focussing on the five high schools (see Figure 2).Data was obtained, with their permission, from a number of sources, that is, the schools, parents, school employees and the Gauteng Department of Education.All participation was voluntary, participants gave informed consent and anonymity was guaranteed.Field workers working in pairs undertook a field audit of the each school's physical resources.The results of which were verified using qualitative semi-structured interviews with school employees.This audit collected information on the libraries, laboratories, number and state of the toilets and the like.The parents completed a questionnaire, distributed using random cluster sampling and adhering to class selection methods based on the work of McLafferty (2010) and Kumar (2005).The questionnaire asked questions of a categorical nature with fixed responses.The questionnaire requested data on the following: (a) residential information (b) family information (c) school choice decisions and (d) costs of schooling.
The study has a number of limitations.The parental questionnaire return rate was 45% (73 returned), some of the questions, in particular questions relating to household income and occupational status, were poorly answered or misunderstood, and so these were not included in the analysis.This is a shortcoming of using self-completion questionnaires.In future, such questions will have to be carefully piloted to ensure parents understand them and complete them.This should improve data validity.Two of the high schools elected to drop out of the audit process midway.Some schools only reported on their physical resources, field workers were not allowed to view them.Lastly, there is much criticism of using matriculation scores as an indicator of educational success (Carnoy & Chisholm, 2008).It is argued that matriculation scores lack validity as some schools either encourage weak learners to repeat Grade 11, encourage weak learners to drop out of school prior to completing Grade 12 or urge learners to substitute a difficult subject with an easier one.Furthermore, learners can obtain a matriculation pass without passing all their subjects.Lastly, some academics argue that even the pass mark is a problem, because the requirements to obtain it are too low (Gilmour & Soudien, 2009).Thus, the matriculation pass mark may not be a true reflection of achievement.Nevertheless, for parents and learners a matriculation pass carries a great deal of weight and it is certainly an essential 'passport' to any form of tertiary education and most jobs (Bhorat, 2004;Hunter 2010).

Results
On average, the high schools have 32 classrooms and 1611 learners each, with 51 teachers, nine support staff members and four administrative staff members per school.Virtually all the learners are black, with only 2.8% classifying themselves as either Asian or Coloured.Almost all (95%) of the learners reside in Alexandra itself, with a few commuting into Alexandra each day from the surrounding suburbs of Marlboro, Lombardy East and Rembrandt Park.There was some confusion as to the quintile 2 status of the schools.According to the school management, all of the schools were designated either Quintile 1 or Quintile 2 schools.According to the GDE, three are Quintile 2 schools, one is a Quintile 3 school and one is a Quintile 5 school.In Gauteng, Quintile 1 schools have been non-fee schools since 2007 and Quintile 2 schools since 2010 (Fleisch & Woolman 2004;Fleisch 2008;Døssing et al 2011).Quintile 3 schools have more recently also being declared 'no-fee' schools.The study found that despite no school fees being charged, education was not free.Some 17% of parents reported paying 'donations' in the region of R500 per annum to the school [although the schools themselves reportedly 'requesting' annual donations of less than R100].Parents indicated that they also bear additional school costs, with some 93% saying they paid for school uniforms, 54% bought school stationery, 41% paid for extra-curricular activities, 23% bought school note books, 8% paid for school sports and 7% purchased school textbooks.

2
The Quintile System, has been used since 2007, to allocate government subsidies to schools.Public schools are ranked according to their socioeconomic profile, which is based on mapping the 2001 census data.The poorest schools are ranked Quintile 1 and the wealthiest Quintile 5 (Bell & McKay, 2011).Some parents (10%) also had to find funds for a commute to school.Of these, 4% said they paid R200 a month, 3% paid between R201 and R400, and another 3% paid between R401 and R800 a month for transport (Bell, 2009).This finding on the school commute cost is supported by many other studies (NDE 2003;Fleisch & Woolman 2004;Ndimande 2006;Redpath 2006).Thus, such 'hidden' costs of schooling may be further impoverishing households.
Access to safe, hygienic ablution facilities are crucial, both to control the spread of disease and the vectors of disease, as well as for personal dignity (Abrahams et al 2006).The audit found that all of the high schools had flushing toilets, but the toilet-to-learner ratios varied.Some had relatively low toilet-to-learner ratios, such as 1 toilet to 40 learners, while another had a ratio of 1 toilet to 53 learners.One had a shocking ratio of one toilet per 160 learners.Such ratios directly impact on the ability of a waterborne sewage system to cope.High ratios usually result in significant maintenance problems.In two of the schools, the toilets were filthy, odious and infested with flies.It is highly likely that such sanitation problems actively discourage learners particularly, girls, from attending school at times.In general, the male toilets were in a worse condition than the female toilets.Although some schools provided the learners with toilet paper, none provided soap.However, a high ratio of learner to toilets did not preclude one of the schools from keeping the toilets in good condition, providing the learners with soap and toilet paper and ensuring that there are no broken toilets.Thus, management of facilities by the school also matters.
Academic success is associated with academic literacy, which is strongly linked to a culture of reading.To foster a culture of reading, children need access to books.A way of accessing books is to ensure that every school has a functioning library (Hart 2013).This study found that while all of the high schools had libraries, none had resident librarians.Furthermore, the library was generally small and the number of books limited.One school had a library in name only, as it was essentially a storeroom for prescribed school textbooks.Another school had library books that were very old and out of date.The purchasing of library books was found to be less than ideal.One school never bought books, another one only every three years, a third one bought books on an ad hoc basis.Thus, only one high school could be said to have a functioning library, that is, a library with sufficient books and system that allowed learners to borrow them.For most Alexandra high school learners then, exposure to the written word is limited.
The South African government has continually reiterated the need for the country to generate more science graduates.This can only be achieved if more learners are encouraged to take science at school level.One of the many reasons why learners shun science, however, is that pass rates at matriculation level are low, making it a 'risky' subject to take.As science is conceptually difficult to grasp, conducting experiments is generally considered one of the best ways to enable learners to understand scientific concepts.To conduct experiments and comply with health and safety standards, a science laboratory has, therefore, long being considered an essential educational resource (Hofstein & Lunetta 1982).This study found that two of the high schools had no science laboratory at all.For the rest, only one of the laboratories was operational.That is, it had equipment, chemicals and a science teacher, who regularly used it.In one high school, the science laboratory was permanently locked and dust coated the desks and the solitary microscope.
The current economy places a great deal of value in being computer literate.Being in possession of Information Technology related skills also makes a person more employable.
Thus, there is a direct need to expose learners to computers while in school if such learners want to achieve success once they have left school (Bovée et al 2007).This study found that only two of the high schools had a functioning computer laboratory fully connected to the Gauteng Department of Education's Gauteng Online Project.In these two schools, learners had daily access, both during school and after hours.For one school, however, the computer facilities had been vandalised and then stolen.Another had no computers at all.Thus, not all learners in these high schools will attain computer literacy.The matriculation pass rates for the Alexandra schools are low relative to the rest of the education district [Johannesburg North] they are located in and to the average matriculation pass rates for Gauteng (see Table 1).Only two schools had matriculation pass rates of over 70 % in 2011.Nationally, any school with a pass rate of below 60% is targeted for intervention, making all but one of these schools to be on average, poor performers even on a national level.For Gauteng, in particular, where overall matriculation pass rates are generally high, these scores are extremely low.Of concern is that the 2011 results were a lot better than those for 2010.While the improvement in 2011 may be the fruits of a well-planned intervention, it may also be a result of learners being held back in Grade 11, being moved to less challenging subjects or, worse, the weakest ones simply dropping out of school.Alternatively, perhaps the results for 2010 were artificially low because of the interruptions relating to the World Cup and the teachers' strike.It is more likely, however, that the high learner to teacher ratio and the low level of physical resources is structurally inhibiting matriculation success, in line with the findings of Gustafsson & Patel (2006).Importantly the ratio of teachers to learners of 1:32 is much higher than the 1:24 teacher to learner the Pienaar & McKay (2014) study demonstrated to be necessary if high matriculation pass rates are to be achieved.Learners attending these schools have less of a chance of obtaining their matriculation certificate and, as such, will not be able to enter tertiary education and will find the labour market extremely difficult to access (Hunter, 2010).As such, their socioeconomic status is unlikely to improve.Importantly, additional research should be undertaken into High School D so as to find out how this school manages relatively good results.Such may inform future policymaking.

Schools
Considering the poor quality of the physical facilities and the low matriculation pass rates, it is unlikely that such schools would be 'schools of choice'.Indeed, the literature abounds with evidence that, if black learners were to have a choice, they leave township schools for either suburban or private schools (Sekete et al 2001;Soudien 2004;Fataar 2007;Soudien 2010).
But other studies have revealed that only a small percentage of Alexandra learners attend schools outside of the township.For example, the Bell & McKay (2011) study found that Alexandra learners constituted only 8% of the Sandton school population and the Machard (2014) study found Alexandra learners constituted only 4% of Johannesburg inner city private school population.Thus, it seems that the majority of children living in Alexandra attend local schools.Why then do parents send their children to such schools?When asked, the overwhelming majority of parents indicated that they self-excluded themselves from betterresourced schools due to a lack of household finances.Close on 38% said they could not afford the school fees of the resourced schools, 25% said they could not afford the commute costs to these schools and some 19% indicated that other school-related expenses (such as school uniforms) prohibited enrolment in neighbouring resourced schools (Bell 2009).None indicated that they knew about the possibility of a fee waiver for their children to attend these neighbouring high-fee schools.Thus, for many, there is no real choice in terms school enrolment.

Discussion
Alexandra schools have poor facilities, especially when compared to neighbouring ex-Model C schools.Crucial education-related physical resources such as libraries, laboratories and computer facilities are generally absent, in a poor condition or poorly utilised.This is similar to what Lemon (2004) Lemon & Battersby-Lennard (2009) found for Cape Town.Thus, the apartheid resource backlog has not been addressed and these schools continue to resemble apartheid ones (Case & Deaton 1999).Unless the provincial government takes responsibility for fully rectifying the infrastructural backlog, the educational deprivation of learners in Alexandra will continue.Thus, the study welcomes the promulgation of Norms and Standards for School Infrastructure by the Minister of Basic Education in 2013.Such standards will now enable parents to legally demand the physical upgrading of their schools.In addition, the state of the toilets shows that, school management is also a problem, a finding that supports the work of Gustafsson & Patel (2006) and van der Berg (2008).Thus, human capacity in terms of school management systems needs to be built, as the toilet situation demonstrates that money alone is not the solution.The study also confirms the findings of Pienaar & McKay (2014), that some schools are incorrectly assigned their quintile ranking and so are correct to dispute their quintile rankings.Overall, local level policy making may have to take a decision to allocate additional funds are allocated to these township schools, not only for maintenance and infrastructure but also to employ more teachers to address the high learner to teacher ratio.Furthermore, this study confirms confirming the findings of Gustafsson and Patel (2006) that there is a 'hidden' cost to schooling due to the cost of school uniforms, 'donations' and a commute.

Recommendation
Funding for school infrastructure upgrading will not be easily sourced.Perhaps schools can enter into public-private partnerships to address some of these needs.The burden of school uniform purchases could perhaps be reduced by donating or recycling school uniforms.The cost of the school commute could be mitigated by donating bicycles and setting aside parts of the road as dedicated cycling lanes, or whole roads as pedestrian and bicycle friendly routes.The issue of schools requesting 'donations' needs investigation by the relevant educational authorities as it may be linked to the dispute over quintile rankings and these donations may not be voluntary.

Conclusion
This case study demonstrates that inequality in South Africa also extends to education.While former racial discriminatory practices are partly to blame, the failure to allocate funds to upgrade geographically marginalised township schools is exacerbating the problem.This study reveals the challenges associated with the local level education policies pertaining to quintile ranking, which whilst progressive in intent, may not be so in implementation.Poor quality education is increasing the rate of unemployment and preventing learners from accessing higher education, demonstrating that scholars such as Lemon (1995), Fataar (1997), Weber (2002) and Soudien (2007) who warned that the apartheid education backlog would have profoundly negative effects on educational outcomes, were correct.Poor Alexandra parents are unable to enrol their children in resourced schools due to financial constraints, regardless of their legal rights and geographical proximity to such schools.In addition, the study reveals that school attendance comes with costs most parents can ill afford, due to the cost of the commute, school uniforms and other education related expenses.The long-term quality of life prospects of learners enrolled in these township schools are dim and as such, they are forming a pool of people structurally excluded from the formal economy.It is argued here that the combination of high cost (relative to incomes) and poor educational outcomes is working in tandem to lock Alexandra learners into a 'truly disadvantaged' or underclass position of intergenerational poverty.These learners have little hope of achieving upward class mobility, primarily because access tertiary education and/or the labour market is mostly determined by the quality of the matriculation result obtained (Bhorat, 2004;Hunter, 2010).As such, racial and class inequalities are being structurally reproduced in Alexandra and state intervention is required to reign this process in (Soudien 2007;Gilmour & Soudien 2009;Bloch 2010).No action by the state in this regard means that the vicious cycle of underclass formation will gain momentum, the size of the underclass will increase and levels of deprivation deepen.This may begin to threaten the stability of the democratic South African state as marginalised people seek to right these wrongs by taking measures into their own hands.
Note: Racial categorization was embedded into South African society by past segregation and apartheid policies.Their continued use post 1994 is due to this legacy practice, but also as a means to track the degree to which South African society is evolving towards a non-racial society.Use of racial categories in this paper does in no way endorse racism or the use of such categories for racial discrimination or 'labelling'.

Figure 2 :
Figure 2: Map showing the location and distribution of Alexandra schools

Table 1 : Alexandra High School matriculation pass rates (source: GDE, 2012)
found in the Eastern Cape, to what the PIRLS 2006 study found nation-wide (see Howie et al 2008), to what can be deduced from National Department of Education statistics (see Dubbelman 2011) and to what It is recommended that a number of research avenues pertaining to education in Alexandra Township are explored in future.This includes finding out where the learners who did not complete high school are; if any Alexandra learners are in private schools other than in the Johannesburg Inner City, if they have enrolled in public schools outside of Sandton, as well as exploring to what extent are Alexandra learners able to access tertiary education and the job market.This will help provide a more nuanced view of the class mobility prospects of Alexandra residents.