Azra Hoosen | ah@radioislam.co.za
23 August 2024 | 16:30 CAT
2 min read

Stats SA is facing a significant crisis after it decided not to release crucial data from the Census 2022, specifically relating to employment and income. The statistical agency has acknowledged that certain portions of the census data are unreliable and therefore unusable. Concerns about the integrity of the data were first raised last month by Professor Tom Moultrie, a demographer from the University of Cape Town (UCT).

Speaking to Radio Islam, Professor Moultrie explained that the situation is complex, clarifying that the issue isn’t about withdrawing data but rather a decision by Stats SA not to release crucial information.

According to Prof. Moultrie, redoing the census or specific parts of it is not an option, he said that Stats SA has maintained that the Quarterly Labour Force Survey can be used to obtain unemployment data, which is the foundation for all headline unemployment figures in the country.

While Stats SA has acknowledged issues with certain parts of the census data, including income, employment, and fertility, they assert that the overall population estimates remain reliable. However, Prof. Moultrie argue that inaccuracies in the census could contaminate other surveys, leading to unreliable data across the board. These surveys, which rely on census data as a sampling frame, are crucial for measuring employment, wealth distribution, and poverty levels. Concerns have been raised about the potential knock-on effects on budget allocations for municipalities and provinces, which are based on population distribution and income data.

“The scale of the data which is available in the census, means that one can drill down to far smaller areas and micro-suburbs to understand income and employment dynamics, which we won’t be able to do now,” he said.

Stats SA plans to use the census data to create a new Master Sample Frame for future surveys, despite concerns from researchers about the accuracy of the information. The Statistics Council has also expressed confidence in the overall quality of the census but advised caution regarding specific data points.

“Given the concerns, and it was clearly a very brave move to try and run a census during the Covid-19 pandemic period, we don’t know the full scale of what went wrong. All of this, our next Census, we cannot afford to make another mess. Therefore, it is vitally important to really unpack what went wrong to avoid those mistakes with the next one,” he said.

The controversy comes as Stats SA prepares for the 2030 census round, with plans to hold an expert meeting in September to address the challenges faced during Census 2022, which was conducted during the Covid-19 pandemic.

“At the moment we have very grave concerns. Stats SA has said they will release to the public and researchers about 10% sample of the date, about 6-million records, which will allow us to get a better picture other than what they have already put in the public domain so far at which point we will give commentary of whether we think the results of the census are plausible,” he said.

The Treasury is currently engaging with Stats SA to ensure that budget allocations remain fair and reflective of diverse needs.

LISTEN to the full interview with Ml Sulaimaan Ravat and Professor Tom Moultrie, a demographer from the University of Cape Town (UCT), here.