Census 2001’s ultimate success depends on what the government does with the results, writes Evidence wa ka Ngobeni
On Wednesday September 10, 85 000 workers from Statistics South Africa (Stats SA) will fan out across the country for the start of Census 2001, with only three weeks to record the details of more than 10-million households.
Census 2001 the second post-apartheid census will provide a tool with which to measure progress and help the government to plan policies intended to alleviate poverty.
Stats SA puts it like this: “Imagine a world without statistics. Investors will not know where the growth in the economy is. Governments will not know where to build roads, clinics or new houses. It will be impossible to plan.”
More crucially, how will our policymakers know whether their policies have achieved their objectives? It is against this background that one does not object to the use of taxpayers’ funds (about R642-million for Census 2001) to run the census.
However, this view can be sustained only if the results of the census are being used effectively by the government to develop and monitor the growth of our country.
So questions arise. Is our government and its policymakers responding to the census results? Do our political leaders understand the importance of the census or its impact on service delivery?
Is there a coordinated effort to ensure that the census is being exploited to benefit the unemployed, the homeless and poverty-stricken South Africans?
The answers to these questions are regrettably murky. Stats SA, which falls under the Department of Finance, has publicly stated that the government’s use of statistics is not adequate.
However, the extent of that inadequacy is not yet clear. Head of Stats SA Pali Lehohla does not describe the size of the problem, but he does slot in some perspective. African countries, he says, are still grappling with the issue of census, and South Africa is not immune.
In a telephone interview, Lehohla said Stats SA needs to articulate the statistics it provides to the government. In the past, Lehohla said, “we have been communicating the census but not articulating” and “we need to start doing that”.
What can one make of this? It raises questions about capacity within the the government to handle statistical information.
University of Pretoria Professor Piet de la Rey says the government has underestimated the impact of HIV/Aids when handling statistical information.
“The government needs to understand that because of HIV/Aids the development needs (which are highlighted by the census) of the country will change dramatically. The government needs to adjust its budget, taking into consideration the HIV/Aids issue, but that has not been happening. In fact we have seen budgets going upward, not downward.
“Say there is a need for 20 000 houses at present; a few years down the line, that could change, because a lot of people have been wiped out by HIV. Similarly, you can say there are 20 000 houses is the country while, because of HIV, most of them are standing empty.”
The Census 96 report issued in October 1998 was hailed as an important development resource and provided the country with the first baseline data against which to measure change.
For the first time after the 1994 general election, South Africans were given an insight into the extent of glaring socio-economic disparities.
Illiteracy, poverty, a high unemployment rate (it stood at 33,9%), lack of basic utilities such as toilets, water and sanitation systems were highlighted by the report.
Although Census 96 ignored the mortality rate, this year’s census will include it. Census 2001 will also look at the pattern of deaths.
But there will be problems: “HIV/Aids,” says De la Rey, “is not a notifiable disease and people will not come out to say their relatives and family members have died from it. Therefore, the statistics may be unreliable.”
In his speech to Parliament, Minister of Finance Trevor Manuel gave examples of why South Africa needs to go the route of census.
“In 1996, 64% of households in the Northern Province were using wood as their main energy source for cooking. Has this situation changed? If so, how? Therefore, a successful Census 2001 … will inform planning, monitor progress and measure performance and impact of endeavours.”
Manuel’s speech provided an insight into how some government departments were using the census statistics: planning the fiscal, elections and demarcation processes.
What does not come out of Manuel’s speech, however, is whether the government has the capacity to respond to statistical information.
The question of capacity is significant. During the apartheid era the results of the census, conducted by the now defunct Central Statistical Service (CSS), were dismal, to say the least.
The fact that the CSS’s census was characterised by, among other things, grossly inefficient systems, late reporting and passing on of irrelevant data is well documented.
Since its formation the new body, Stats SA, has made some significant achievements to repair the damage caused by the CSS and improve the quality of census. The CSS used a demographic model, rather than an actual count, for example, to estimate the size of the population.
But there are still a lot of challenges. As Lehohla puts it: “The current census does not do the analytical work. That is the work of government departments.”
Meanwhile, De la Rey says that Stats SA itself “needs to build capacity in order to deal with the emerging dynamics”.
Lehohla echoes this view. He said: “We need to look at the way we collect our data, the issues around relevancy, accuracy and how often we conduct the census”.
The law requires that a population census be conducted every five years but with census takers poised to go into action, there are still issues that need to be addressed, seen within the broader challenges facing the civil service.
Lehohla’s admission that the use of census by the government to inform social policies is not “inadequate” demonstrates the extent of the challenge, but his position seems clear: ignoring the census could have serious repercussions.
The census, for example, will offer detailed statistics on how many hours it takes for residents of one area in the country to travel to another area. This data can help the government to develop a responsive and informed public transport system.
But more crucially it can be used to map out disaster areas and help to identify potential dangers. The problem of shack fires, for example, could be detected before they occur.
That makes it more worrying that the relevant local authorities, fire departments and other disaster management authorities “are not using the census well”.
One of the key mandates of the post-apartheid Stats SA is to produce accurate and accessible statistics to help advance, among other things, economic growth and democracy in the new South Africa.
New logistical initiatives, such as IT technology, have been introduced, and scanners will be used to read the census forms that members of the public have filled in, reducing the time the country will wait to receive the final report.
But the major challenge facing Stats SA is to get people to participate in the census. Stats SA has put aside about R40-million of its R642-million to “inform the public” what the census takers in their bright yellow T-shirts are up to.
There is no doubt that more innovations to get overwhelming public support on the part of Stats SA will become a key leverage point for integrated national and economic planning.
Lehohla agrees that his organisation needs to work harder. “We need to focus on building a formidable statistical system, and engage in vigorous training of our staff.”
The Census 2001 presents an opportunity to do just that. The result will surely be heartening: better professional monitoring and evaluation of service delivery programmes by national and provincial government departments.