contemporary urban Johannesburg-Soweto
5.4 Nearest school based operationalization of learner mobility mobility
The final approach to measuring learner mobility involves determining whether or not children are enrolled at their nearest grade-appropriate school.
As discussed in Chapter 3, while this is not a perfect indicator of engagement in school choice, the proportion of children not attending their nearest school is expected to provide a fair approximation of the proportion of children engaging in choice. Again, due to substantial similarities over time, the data for 1997 and 2003 is presented together.
The first key finding using this approach to measuring mobility is that less than 20% of children are actually attending the grade-appropriate school nearest to their homes in both 1997 and 2003 (see Table 5.8 below). This figure is surprisingly low, and suggests that over 80% of children are travelling further than strictly necessary in order to attend school. One possible reason that children might not be attending their nearest grade-appropriate school could be that the school in question is an independent (private) school. Due to this, two
138
sets of figures are presented, one including only public schools, and one including independent schools as well. As is clear from the data in Table 5.8, this makes very little difference to the results.
However, when the 2003 data is disaggregated by schooling phase – that is, when children enrolled in primary school are separated from those enrolled in secondary school – an interesting pattern is revealed. Despite hypotheses that mobility should be higher amongst high school children, a substantially higher proportion of these children are attending their nearest school (just under 22%). The overall proportion of children attending the nearest school remains the same because the proportion of primary school children attending the nearest primary school actually falls fairly markedly to just over 15%. While the higher proportion of high school children attending the closest school may just be due to a smaller number of available high schools (see Chapter 4), the lower proportion of primary school children attending their closest school at age 13 is more intriguing. One potential explanation is that children who are attending schools further afield perform more poorly, making them more likely to still be in primary school at age 13. An alternative, and somewhat more plausible explanation may be that when children fail a grade, their parents are more likely to try sending them to different schools, which may be further
139
Table 5.8: Number and percentage of learners attending the school closest to their home in 1997 and 2003, and the mean and maximum distances to the schools nearest to sample members’ homes
The data on the distance from children‘s homes to their nearest schools provides an additional interesting finding: the mean distance a child needs to travel to attend their nearest primary phase school is just approximately 400m, and less than 5% of children need to travel over 1km. When contrasted to the actual distances children are travelling – previous calculations indicated over 50% of children travelling over 1km – this highlights the extent to which travel, even of moderate levels, appears to be due to children attending schools further from home than is strictly necessary.
In 2003, however, not all children are still within easy walking distance of a grade-appropriate school. Although 95% of children in 2003 have to travel less than 1.15km to reach their nearest school, there are a small number of children who have to travel over 3km. This is probably primarily due to the fact, discussed in Chapter 4, that there are substantially fewer high schools in the Johannesburg-Soweto area, due to their typically having a somewhat larger
140
size. Disaggregating the children by schooling phase supports this hypothesis, as the data generated for the primary school children remains similar to that generated in 1997. Nonetheless, even when children are in high school, the average distance travelled remains substantially greater than the distance a child would need to travel to access his or her nearest school.
5.5 Conclusion:
This chapter has explored three different approaches to defining and measuring learner mobility, and provided data about the extent of learner mobility in Johannesburg-Soweto on the basis of each of these definitions. Each definition is likely to prove particularly valuable for certain purposes, and in certain contexts. Using a distance-based measure provides both a binary and a continuous measure of mobility. The distance-based binary measure is of the form that is typically used in school choice related policy, and is therefore particularly valuable in assessments of the appropriateness or applicability of policy. The continuous measure of distance is particularly useful in exploring the actual extent of mobility and what it entails for particular learners in terms of the investments they are required to make, both financially and in terms of time. In addition, it allows for the examination of the distribution of the distances travelled by the entire sample, and, as it is the measure that has been most commonly used in the existing literature, it also allows for comparison with previous findings.
The definition of mobility based on census geography is particularly useful in that it makes use of generally accepted geographical areas to explore the extent to which mobility is occurring within and between these areas. This is helpful in identifying whether learners are travelling between areas historically designated for different race groups, and thereby significantly enhancing the quality of education they are likely to receive. Additionally, it is, and thus identifying those learners who are likely to be making the most substantial
141
economic investments in their education. Finally, the definition based on whether or not the learner is attending the age-appropriate school closest to his or her home is useful in highlighting the extent to which even learners with relatively low levels of mobility may be engaging in more travel than strictly necessary or anticipated, and may also be engaging in school choice, particularly within the historically disadvantaged areas.
This chapter has made two key contributions to the literature. Firstly, in providing three different approaches to the conceptualization and measurement of learner mobility, it has significantly enhanced the methodological tools available to the study of this practice. Secondly, it has, for the first time, provided population-based data on the extent of learner mobility in contemporary urban South Africa. In so doing, it has identified preliminary evidence to suggest that there may in fact be two patterns of school choice and mobility in operation in Johannesburg-Soweto. Firstly, there is a group of approximately 25% of the sample who are engaged in substantial travel from home to school on a daily basis, and who seem likely to be making significant investments in this mobility. Secondly, and somewhat less expectedly, there is also evidence that a large proportion of children who are not travelling substantial distances to school are still engaging in mobility and school choice.
Even though they are attending schools relatively close to home, they are not attending the nearest grade-appropriate school to their home, and are often travelling to schools that are not located in the same residential areas as their homes. These patterns, and their importance to understanding the implications of learner mobility to educational access and equality, are explored in greater detail in the subsequent chapters.
142