Quality of life disparities within wards in Gauteng

Introduction

Inequality is an enduring characteristic of the South African landscape. The country ranks among those with the highest inequalities globally on socioeconomic dimensions. The inequalities found in South Africa have been studied in terms of differences in access to services, economic opportunities and wealth outcomes between males and females, population groups, urban-rural regions, provinces, and geographic areas within municipalities such as townships versus suburbs.

Katumba and Culwick-Fatti (2015) have previously analysed the spatial distribution of Quality of Life (QoL) index values at the ward-level using data from the Gauteng City-Region Observatory’s (GCRO) Quality of Life (QoL) Survey. This provides a view of spatial disparities in quality of life across wards. Using data from GCRO QoL 7 (2023/24) Survey we update this perspective below, confirming enduring inequality in quality of life across the province. However, in this Map of the Month, we add another level of understanding by analysing the QoL Index differences among individuals within wards. Our analysis first considers the average QoL Index scores of the bottom 25% in each ward, then the average QoL index scores of the top 25%, and then the differences between these averages to derive a measure of in-ward disparities in quality of life. We round off by briefly discussing possible reasons for the patterns of inequality observed.

Quality of Life Index

The GCRO QoL Index is a multidimensional score of wellbeing scaled to range from zero to 100. A QoL Index score of zero represents the worst condition, and 100 means a perfect quality of life. The steps followed to calculate the GCRO QoL Index are detailed in Katumba et al. (2022). The Index is based on 33 objective and subjective indicators of individuals and their households’ wellbeing, and are pooled into seven dimensions, namely health, life satisfaction, government satisfaction, participation, safety, services and socioeconomic status. The dimensions and indicators have different weights on the resulting QoL Index (Naidoo et al., 2024). In the QoL 7 (2023/24) Survey, the average QoL Index for Gauteng province was 59.6 out of a possible 100 (Naidoo et al., 2024). However the QoL Index score differs between each surveyed respondent, and in turn mean scores differ across population groups, dwelling types, levels of education, and different geographical parts of the province.

Figure 1 illustrates this spatial differentiation. Updating Katumba and Culwick-Fatti’s (2015) analysis with the latest QoL data it displays the average QoL Index score for each ward across the Gauteng province. The lowest average score for a ward is 37, and the highest average score is 78. The map clearly shows that the central parts of the province have significantly higher average QoL Index scores. Most of the wards with the highest average QoL indices are found in the City of Johannesburg (the areas north of the Riverlea place marker) and the City of Tshwane (the areas around the Irene place-marker). By contrast, most townships and many peripheral areas have much lower average QoL Index scores.

Figure 1: Average QoL Index score by ward. Data source: QoL 7 Survey (2023/24).

Spatial patterns of QoL inequalities within wards

While Figure 1 confirms a known pattern, the analysis in this Map of the Month delves deeper. Here we examine the spatial patterns of ward-level inequalities in QoL Indices not by looking at inequality between wards, but rather at inequality within wards. This is done by computing the average QoL Index score of the bottom 25% of QoL Index scores in each ward, then the average of the top 25%, and subtracting the two to get a percentage point difference. The rationale for using the bottom 25% and the top 25% is to identify inequalities whilst controlling for individual respondent outliers who may have very high or very low quality of life. However, this does not mean that outliers do not represent the lived experiences of some residents in Gauteng. A large gap indicates significant disparities, suggesting that some individuals may be experiencing much better or worse conditions than others in the same ward. These variations help highlight where inequalities exist, providing a clearer picture of where targeted interventions may be needed to improve the conditions of those with lower QoL.

Figure 2 below shows the average QoL Index score of the bottom 25% of individual respondent QoL Index scores per ward. The deeper the shade of red, the lower the average index score from the bottom 25%. The lowest ward average is 20, and the highest ward average is 71. Wards in the periphery of Gauteng rank among those with the lowest average QoL Index scores for the bottom 25%, for example around Hammanskraal, Rayton, Springs and Carletonville.

Figure 3 plots the average QoL Index scores of the top 25%. The deeper the shade of red, the higher the average for the top 25%. The lowest average score is 52, and the highest is 88. The map shows that wards with the highest average QoL index scores for the top 25% concentrated mainly in the central areas of the province. However, there are also wards in Rayton, Laudium, Lenasia, Vanderbijlpark and elsewhere that rank amongst those with higher – in the 79.8 to 85+ bracket – average QoL Index scores for the top 25%.

Figure 2: Average QoL Index scores for the lowest 25% in each ward. Data source: QoL 7 Survey (2023/24).

Figure 3: Average QoL Index scores for the highest 25% in each ward. Data source: QoL 7 Survey (2023/24).

In our analysis the difference between the average for the bottom 25% and the top 25% denotes the magnitude of QoL inequality in a ward. The higher the percentage point difference between the two averages, the higher the inequality. The percentage point difference per ward is shown in Figure 4. The wards with the greater QoL inequalities are denoted by darker shades of red. The smallest percentage point difference is 10, and the largest is 48.

Figure 4: Percentage point differences in the average QoL Index scores of the lowest and highest 25% per ward. Data source: QoL 7 Survey (2023/24).

Although the pattern is not uniform, Figure 4 suggests that inequality in quality of life within wards tends to be smaller in the wealthier central suburbs of the province. It is noteworthy that the central parts of the Gauteng province are mostly suburban areas with greater proportions of high-income households – these show up as having the highest overall quality of life in Figure 1. In turn, disparities in quality of life are larger in low-income areas, many on the periphery of the province. Bands of relatively higher within-ward inequality are visible on the far western, southern and north-eastern edges of the city-region. Illustratively, wards around Bronkhorstspruit, Impumelelo, Ratanda, Hammanskraal and Carletonville have QoL inequalities in the ranges above 23.3-28.9 percentage points and above.

In the City of Tshwane, wards with notably higher percentage point differences are in areas like Rayton, Laudium and to the north of Theresapark. Here, those in the top 25% are, on average, some 35 percentage points better off than their counterparts in the bottom 25%. In the City of Johannesburg, wards with high disparities can be found in places like Little Falls, north of Jukskei Park, and areas between Riverlea and Lenasia. In Ekurhuleni, wards around Vosloorus, Springs and Germiston have the highest levels of inequality between the average QoL Index scores in the top and bottom 25%.

Discussion

Ward-level inequalities in QoL are a reality in Gauteng, and exist in all types of residential locations. There are different ways of interpreting the existing QoL inequalities across the province. In this section we consider possible reasons for the ward-level QoL inequalities. To this end, Figure 5 extracts 10 wards with the smallest QoL differences and 10 wards with the largest QoL differences. To better understand the primary drivers of ward-level disparities, the averages for the bottom and top 25% across seven Quality of Life Index dimensions (services, socioeconomic status, government satisfaction, life satisfaction, health, and participation) – each ranging from 0 to 10 – were calculated, and the differences analysed for these wards.

Figure 5: QoL inequality in the ten wards with the lowest QoL Index score disparities, and the ten highest. Data source: QoL 7 Survey (2023/24)

Interestingly, in Mohlakeng, Rand West, there is a ward with a very small difference in the QoL Index scores of its top and bottom 25% (Ward 22), and also a ward with very large QoL inequality (Ward 13).

In Rand West Ward 13, the overall QoL disparity is 47 points, primarily driven by differences in services (8 points), socioeconomic status (6 points), life satisfaction (4 points), and government satisfaction (4 points). These differences are related to the settlement context of the ward, where most respondents with high QoL scores live in township houses, but a fair proportion of the respondents in the ward live in informal houses, backyard dwellings, and hostels where QoL scores are much lower.

In contrast, Rand West Ward 22 has a smaller QoL disparity of 14 points, mainly influenced by health (3 points), socioeconomic status (3 points), and participation (2 points). Here, most respondents live in township houses where there is more uniformity between QoL scores.

Despite being located in the same settlement (Mohlakeng), it is clear that parts of the two wards are differently affected by service delivery, socioeconomic status, government satisfaction, and life satisfaction, closely related to the housing composition of the wards. Respondents in formal township housing have relatively higher QoL scores, partly because these houses are relatively well-serviced and possibly because some accommodate higher income earning residents that have remained in townships in the post-apartheid era. These would be respondents who are employed and have more socio-economic stability compared to respondents in the same ward living in hostels or informal dwellings.

In another example, Noordheuwel (Ward 22 in Mogale City) and Little Falls (Ward 85 in Johannesburg) both share an average top 25% QoL of 81 points. However, the average QoL for the bottom 25% is very different - 68 in Noordheuwel and 33 in Little Falls. The primary drivers of the QoL disparity in Little Falls are services (8 points), socioeconomic status (7 points), life satisfaction (4 points), and safety (4 points). Ward 85 covers a large area and includes a variety of dwelling types. While most of the ward includes formal suburban houses and cluster housing, a part of the ward also includes an informal settlement where QoL scores for the service and socioeconomic dimensions are much lower. Inequality within a ward becomes very clear when these vastly different settlements are located in the same administrative boundaries. Hence, ward-level disparities should always be interpreted with this context in mind.

In contrast, Noordheuwel is a ward that mostly includes formal housing (suburban, township and estate housing) where QoL scores are relatively high in all instances. The main contributors to the relatively small QoL disparity in Noordheuwel are government satisfaction (3 points), participation (3 points), and safety (2 points)

In other wards in Gauteng, the reasons for QoL disparities might be very similar to the examples provided above. The differences that are seen in terms of service delivery and socioeconomic status are often caused by the proximity of high-income and low-income households in the same ward - especially where gated communities are located near informal settlements or where employers provide accommodation to their domestic workers in their backyards for convenience. The growing presence of backyard dwellings in townships also contributes to QoL disparities, as these often house lower-income households, amplifying inequalities. Furthermore, some wards in suburban areas comprise old-age homes whose residents fare somewhat worse in QoL indicators (such as income, health, and social relationships) compared to residents in other dwelling types in the same wards. These settlement patterns, combined with household income’s critical role in accessing healthcare, housing quality, security, and leisure, add complexity to interpreting inequality patterns within wards.

Conclusion

Quality of Life disparities exist across the Gauteng province and are related to the settlements context of wards. While suburban areas are generally associated with higher QoL, findings show that some township households achieve similarly high QoL, contrasting with low QoL found in certain suburban and township wards due to the presence of informal housing or hostels. This intra-ward variation underscores the complexity of QoL distribution and the influence of socioeconomic factors beyond geographic stereotypes. Ensuring equitable access to services across all wards could play a significant role in addressing these disparities and promoting greater equality within communities.

References

Culwick Fatti, C. and Katumba, S. (2017). 2015 Quality of Life (QoL) index by ward, GCRO Map of the Month, Gauteng City-Region Observatory, January 2017.

GCRO (Gauteng City-Region Observatory). (2024). Quality of Life Survey 7 (2020/21) [Dataset]. Version 1. Johannesburg and Cape Town: GCRO & DataFirst. https://doi.org/10.25828/v2ky-v879

Katumba, S., de Kadt, J., Orkin, M. and Fatti, P. (2022). Construction of a Reflective Quality of Life Index for Gauteng Province in South Africa, Social Indicators Research, 164, 373–408.

Naidoo, Y., Mahamuza, P. and Naidoo, L. (2024). Quality of life and wellbeing: Findings from the GCRO's Quality of Life Survey 7 (2023/24). GCRO Data Brief 25, October 2024. https://doi.org/10.36634/PJCA9220

Note on methods

This analysis is based on data from the QoL 7 (2023/24) Survey. All analyses were conducted in Stata Version 18.0 and results were exported to Datawrapper for mapping. Datawrapper is an online-based data visualisation tool that interactively presents information in the form of maps, charts and tables. To compute QoL gaps in each ward, we followed four simple steps. In the first step, we sorted QoL indices in each ward from lowest to highest. In the second step, we allocated the QoL indices into four quantiles each containing 25% of ward’s indices. This means the bottom 25% of the scores in a ward were allocated to the first quantile, the top 25% were allocated to the fourth quantile, and the remaining 50% were equally allocated to second and third quantiles. The third step involved computing for each ward the average QoL score of the first quantile (bottom 25%), and separately for the fourth quantile (top 25%). The results of this step are the ones presented in Figure 2 and Figure 3. In the fourth step, we calculated the difference between the averages of the bottom and top 25% for each ward.

Edits and inputs: Graeme Götz, Laven Naidoo, Samkelisiwe Khanyile

Suggested citation: Ndagurwa, P., Miles-Timotheus, S. and Hamann, C. (2024). Quality of Life disparities within wards in Gauteng, GCRO Map of the Month, Gauteng City-Region Observatory, December 2024. https://doi.org/10.36634/QSUT4001