Clusters of dissatisfaction with local government performance
- Chris Wray, Samy Katumba
- Date of publication: 31 January 2016
- Download map
Quality of Life (QoL) surveys, conducted by the GCRO biennially since 2009, reveal an increase in the levels of satisfaction with government-provided services. Specifically the surveys show rising satisfaction between 2011 and 2013 with government provided dwelling, water, sanitation, waste, energy, street lighting, roads, stormwater, municipal billing, cost of municipal services, libraries, public health services, education, and public safety and security. However, at the same time, QoL survey results suggest a steady increase in the levels of dissatisfaction with local government performance.
To understand this disjuncture the GCRO embarked on a spatial statistical analysis of the 2013 QoL survey data. The analysis addressed the following two questions:
- Is there a spatial pattern evident in the dissatisfaction with the performance of local government responses from the 2013 QoL survey?
- What significant determinants may explain the patterns of dissatisfaction with the performance of local government?
To determine whether dissatisfaction with local government might cluster in specific areas, we performed a statistical and spatial test known as spatial dependence. This highlights where similar dissatisfaction levels are clustered together spatially. The diagram below demonstrates how the clustering works, with blocks representing wards. It should be read together with map 1.
Map 1 (below) illustrates spatial clustering in the level of dissatisfaction with the performance of local government on four dimensions of clustering.
- ‘Low-low’ clusters (depicted in blue) represent wards with low levels of dissatisfaction surrounded by neighbouring wards with similar low levels of dissatisfaction with government performance. ‘Low-low’ clusters mainly occur in the central areas in Johannesburg, extending up through the southern and central areas of Tshwane.
- ‘High-high’ clusters (depicted in red) represent wards with high levels of dissatisfaction surrounded by neighbours with similar high levels of dissatisfaction with government performance. These ‘high-high’ clusters are clearly visible in a corridor extending from Vanderbijilpark/Sebokeng in Emfuleni to Bekkersdal in Westonaria. There are also patches of significant ‘high-high’ clustering in areas such as Khutsong, Soweto, and Alexandra.
- On the one hand, ‘low-high’ outliers represent wards with low level of dissatisfaction that are surrounded by wards with high levels of dissatisfaction with government performance.
- On the other, ‘high-low’ outliers represent wards with high level of dissatisfaction that are surrounded by wards with low levels of dissatisfaction with government performance.
The second part of our analysis focused on what factors may spatially explain the concentration of high levels of dissatisfaction in particular parts of the province. The analysis used a technique known as spatial error modeling (SER). A number of key explanatory variables emerged as significant, including dissatisfaction with dwelling, perceptions of a lack of safety, concerns with maintenance, and the belief that politics is a waste of time.
Map 2 (below) illustrates this analysis by showing percentages of respondents dissatisfied with dwelling, one of the most significant determinants of dissatisfaction with local government performance. It is evident from the map that the highest dissatisfaction with dwelling occurred in the Khutsong and Bekkersdal areas in the west; Sebokeng/Vanderbijlpark in the south; central Germiston, Thokoza and Daveyton in the east, Mamelodi, Winterveldt and Hammanskraal in the north; and Orange Farm, Alexandra and Diepsloot in Johannesburg. Unsurprisingly, wards with low levels of dissatisfaction with dwellings can be observed in the wealthier wards in the centre of the Gauteng City-Region.
Overall, results from this spatial statistical analysis suggest that local government needs to focus on issues of housing, safety and maintenance, as addressing these concerns may lead to increased levels of satisfaction in the identified areas.
A more detailed exposition of the analysis and results can be found in the paper: “Spatial statistical analysis of dissatisfaction with the performance of local government in the Gauteng City-Region, South Africa” published in the special issue of the South African Journal of Geomatics, Vol. 4, No. 3, August 2015. The paper won “the best-peer reviewed paper presented at GI 2015” Award at the Geomatics Indaba conference held at Emperor Palace, Kempton Park (South Africa).