Spatial Patterns of Adult Illiteracy in South Africa, 1996 and 2022

Thirty years of constitutional democracy and census-taking have given South Africa a rich statistical record of change. Census data from 1996 to 2022 reveal important gains in education, employment and representation in leadership. These achievements take on particular significance in September, as South Africa observes Literacy Month and marks International Literacy Day on 8 September, a time to reflect on the role of education and literacy in expanding opportunities and shaping people’s lives.

In the absence of direct measures of literacy proficiency, educational attainment is used as a proxy indicator of functional literacy. Using census 1996 and 2022 data, individuals aged 20 years and older with no formal education or whose highest level of educational attainment is below Grade 7 are classified as functionally illiterate, while those who have completed Grade 7 or higher are considered functionally literate. This approach is consistent with international practice of using educational attainment as a proxy measure of functional literacy.

Based on 1996 Census data, a hotspot analysis was performed by Statistics South Africa to examine the spatial distribution of adult illiteracy among South Africans aged 20 years and older. The findings indicated significant spatial clustering, with distinct hotspots and cold spots of illiteracy evident across the country.

What is hotspot analysis?

Hotspot analysis helps us understand where adult illiteracy is concentrated geographically. It does more than identify the areas with the highest levels of illiteracy. It looks at neighbouring areas to see whether high levels of illiteracy occur together, forming a cluster, or whether low levels occur together, forming a cold spot.

For example, if one district has a high level of adult illiteracy but the districts around it have much lower levels, it may simply be an area with a high value. But when several neighbouring districts all have high levels of illiteracy, this suggests a spatial cluster, or hotspot.

The analysis then tests whether these clusters are statistically significant. In other words, it tests whether the pattern is strong enough to suggest that it is unlikely to have occurred simply by chance.

In this analysis, therefore:

  • 🔥 Hotspots are areas where high levels of adult illiteracy are clustered together.
  • ❄️ Cold spots are areas where low levels of adult illiteracy are clustered together.
  • Not statistically significant means that the pattern of neighbouring values is not strong enough to identify a clear cluster.

Adult Illiteracy in South Africa, 1996

The most prominent hot spot of high illiteracy was located in KwaZulu-Natal, particularly in the north-eastern part of the province around Umkhanyakude, where clustering was significant at the 99% confidence level. This represented the strongest concentration of adult illiteracy in the country. Surrounding districts, including areas corresponding to Zululand, Uthungulu (King Cetshwayo), Umzinyathi, and parts of Sisonke (Harry Gwala) and uMgungundlovu, also formed significant hot spots at the 90% and 95% confidence levels. Together, these districts created a broad zone of elevated adult illiteracy extending across much of KwaZulu-Natal.

In contrast, significant cold spots of low illiteracy were concentrated in Gauteng and adjoining districts. The strongest cold spots, significant at the 95% and 99% confidence levels, extended into neighbouring parts of North West, Limpopo, and Mpumalanga. Additional cold spot clusters were identified in Western Cape. These areas were characterised by consistently lower levels of adult illiteracy than surrounding regions.

Large parts of Eastern Cape, Free State, Northern Cape, and sections of Limpopo, Mpumalanga, and North West were not classified as statistically significant hot spots or cold spots. This indicates that illiteracy levels in these areas did not form sufficiently strong spatial clusters, with neighbouring districts displaying more varied patterns.

Overall, the 1996 results show a clear spatial structure in adult illiteracy, with a large concentration of high illiteracy in KwaZulu-Natal and clusters of low illiteracy centred on Gauteng and parts of Western Cape.

Adult Illiteracy in South Africa, 2022

The hot spot analysis based on Census 2022 indicates that adult illiteracy remained spatially clustered, although the extent and intensity of clustering were smaller than in 1996.

The most extensive pattern was a series of cold spots of low illiteracy centred on Gauteng and the Western Cape. The strongest cold spots, significant at the 99% confidence level, were found within these provinces and extended into surrounding districts in North West, Limpopo, Mpumalanga, and Free State. These clusters formed a broad corridor of relatively low adult illiteracy across parts of the country.

Significant hot spots of high illiteracy remained in KwaZulu-Natal and parts of Eastern Cape. The strongest hot spot was again located in north-eastern KwaZulu-Natal, broadly corresponding to Umkhanyakude District, where clustering was significant at the 95% confidence level. Additional hot spots occurred along parts of the KwaZulu-Natal coastline and in districts bordering Eastern Cape, primarily at the 90% confidence level. Parts of the southern interior of Eastern Cape also formed significant clusters at the 95% confidence level.

Compared with 1996, much larger areas of the country were classified as not statistically significant, including substantial portions of Northern Cape, Free State, North West, Mpumalanga, Limpopo, and sections of both KwaZulu-Natal and Eastern Cape. This suggests that adult illiteracy was less geographically concentrated across many parts of South Africa by 2022.

The 2022 pattern therefore reflects a country where low illiteracy was concentrated around Gauteng and Western Cape, while higher levels remained clustered in parts of KwaZulu-Natal and Eastern Cape.

Comparing the 1996 and 2022 Spatial Patterns

A comparison of the two periods reveals important changes in the geography of adult illiteracy.

In 1996, high illiteracy formed a large and highly significant cluster across much of KwaZulu-Natal, with the strongest concentration located in Umkhanyakude at the 99% confidence level. By 2022, this hot spot remained visible but had become smaller and less intense, with the strongest clustering occurring at the 95% confidence level. While north-eastern KwaZulu-Natal continued to stand out as the main concentration of adult illiteracy, the geographical extent of the hotspot had contracted considerably.

Another notable change was the emergence of a more visible concentration of elevated illiteracy in parts of Eastern Cape in 2022. Together with districts in eastern KwaZulu-Natal, these areas formed a corridor of higher illiteracy extending along sections of South Africa’s eastern seaboard.

The pattern of cold spots remained consistent over time. In both 1996 and 2022, the strongest clusters of low illiteracy were centred on Gauteng, with extensions into neighbouring provinces. Western Cape also remained an important cold spot throughout the period. However, by 2022 these clusters covered a broader area, extending further into surrounding districts and reflecting larger regions with consistently low levels of adult illiteracy.

A further difference between the two periods is the increase in districts classified as not significant. In 1996, stronger clustering patterns were evident across several regions. By 2022, many districts no longer formed statistically significant hot spots or cold spots, indicating a weakening of extreme spatial concentrations and a more dispersed distribution of literacy outcomes.

In summary, the comparison shows that the overall extent of high-illiteracy clustering declined substantially between 1996 and 2022. The north-eastern KwaZulu-Natal remained the country’s most persistent hotspot, while Gauteng and Western Cape continued to form the strongest cold spots of low illiteracy. The results highlight both continuity and change in the spatial distribution of adult illiteracy over the period, with regional concentrations becoming less extensive but still clearly identifiable.

ENQUIRIES:  User Information Services   Tel.: 012 310 8600   Email: Info@statssa.gov.za