Statistics South Africa Budget

This premium content has been made freely available

Finance Standing Committee

03 June 2004
Share this page:

Meeting Summary

A summary of this committee meeting is not yet available.

Meeting report

4 June 2004

Chairpersons: Ms B Hogan (ANC) [NA]; Mr S Ralane (ANC) [Free State]

Documents handed out
Statistics South Africa Presentation on Budget
Statistics South Africa Budget Vote 13
Statistics South Africa Strategic Plan 2004-2007 [link to Statistics South Africa website]
Statistics South Africa Annual Report 2002/2003 [link to Statistics South Africa website]
Census 2001: My constituency & Census 2001: Info {link to Census 2001 website]
Statistics South Africa website

Contact for the following documents:
Census 2001: In brief
Statistics South Africa: Stats in Brief 2004

The presentation by the Statistician General outlined the Statistics South Africa value chain, the foundations of the five official statistical surveys, the Statistics South Africa 2004/2005 budget per programme, the key improvements and developmental areas in each of the five groups of statistics, the rate of the customer satisfaction survey, the strategic priorities for the next 12 months as well as the key achievements and future challenges facing Statistics South Africa.

During the discussion Members questioned why the project dealing with Causes of Death had not produced the figures for the number of South Africans who have died of HIV/AIDS and whether legislation was needed to identify this cause of death. Also raised was the concern about determining the Gross Domestic Product figures, the interim measures and shortcomings identified by the Review Task Team on GDP, whether Statistics South Africa was working together with the Reserve Bank and the Revenue Service in gathering the GDP figures as well as Statistics South Africa's plans to adapt its methodology in accordance with international benchmarks. The Committee asked Statistics South Africa to explain the extent to which it was working with Treasury on the issue of administered pricing, the measures put in place to gather information in the deep rural areas, especially the poverty statistics, the status of the National Statistics System and the progress made in completing the transformation process within Statistics South Africa.

Presentation by Statistics South Africa
Mr Pali Lehohla, Statistics South Africa Statistician-General, conducted the presentation (document attached) which outlined the Statistics South Africa value chain, the foundations of the 5 official statistical surveys, the Statistics South Africa 2004/2005 budget per programme, the key improvements and developmental areas in each of the 5 groups of statistics, the rate of the customer satisfaction survey, the strategic priorities for the next 12 months as well as the key achievements and future challenges facing Statistics South Africa.

Mr Patrick Kelly, Statistics South Africa: Service Delivery Improvement, took Members through the new Statistics South Africa website:


Mr B Mkhaliphi (ANC) [NCOP, Mpumalanga] stated that the presentation indicated that 860 086 questionnaires were distributed by Statistics South Africa in 2003 yet, as a Member of Parliament, he never received any of those questionnaires. He thus sort clarity on the sampling process employed by Statistics South Africa in compiling its figures.

Mr Lehohla replied that Statistics South Africa has not yet broken these down and examined the precise number that were completed by businesses, households. etc. It was very difficult to dismember exactly who answered the questionnaires because they were probability samples and therefore not everybody would be included in the surveys. Statistics South Africa was mandated to cover all the areas in South Africa in a manner that was representative of the entire South African population.

Mr Mkhaliphi stated that he was pleased that there was a Statistics South Africa official responsible for provincial liaison as well as district offices, and assured them that the Select Committee would interact with them in future.

Mr Lehohla responded that there were approximately 25 functional district offices, but this number would probably have to be doubled to about 54. They would serve a very important purpose both for collection purposes and for dealing with inquiries on the spot, and would provide Statistics South Africa with additional intelligence with regard to understanding the idiosyncrasies in the supply environment.

He agreed that there was a major backlog in this area and Statistics South Africa had devised a programme to address this. One such example was the Limpopo Province in which the census data was processed such that it became a tool of engagement with local authorities. It yielded definite benefits, and it was hoped that the same would be done in other provinces because it provided a comparable template against which discourse and planning could take place.

Mr Z Kolweni (ANC) [North West] noted that the President had stated last year that he did not have the number of South Africans who have died of HIV/AIDS. This information must be compiled as soon as possible.

Dr Liz Gavin, Statistics South Africa DDG: Statistical Support and Informatics, responded that this would be an important opportunity to explain the causes of death project so as to provide Members with an honest answer as to what can be expected from the outcome of that project. The project involved the use of death notification forms which were completed by medical practitioners from the years 1997 to 2003. The Department of Home Affairs would then take receipt of these forms and the data processing centre would firstly sort and sequence them, and they would then be allocated a number which allowed Statistics South Africa to arrive at a figure of the number of certificates it would have to deal with. The latest estimate stood at approximately 2.7 million forms, but this number would only be verified once this first phase was completed. The 1997, 1999 and 2001 years were however prioritized.

The next stage would be to submit the forms to a group of people who then code them. The particularly crucial portion of the coding exercise would be to look at the information on the immediate cause of death, which was completed by the doctor. There was also space on the form for the doctor to give the additional underlying cause of death as well as any other factors which could have contributed to death. This information was coded in accordance with IDC 10, an international standard for classifying various diseases, and this was captured by data capturers. This would be tabulated into numbers of death according to the various causes, by gender, age and geography by province.

Turning this data into information that non-experts can understand involved two steps: the first would be to look at the groups of information and to deduce the most probable underlying cause of death. In this area epidemiologists would refer to hazard probability models, and essentially involved the factoring in of the age, gender and various other causes listed. Once this was completed the underlying causes of death would then be linked with demographics.

One of the major constraints facing the project was that the people needed for the coding process was a very scarce skill in South Africa, and this resulted in the major bottleneck in the production rate of the figures. The second step involved people with expert knowledge and proven ability in verifying the cause of death before it is captured digitally. Statistics South Africa thus lacked the kind of resources it needed for the project to run as fast as possible. Local companies especially medical aid companies have been contacted to provide this service, and international companies as well as the World Health Organisation. The aim over the long term was to arrive at an automated system that would determine the causes of death, whereas this was currently being done by individuals who used their own knowledge and judgement in determining the cause. The automated system was preferred because it would not result in any sort of bias, and the rules employed by the system would be cleared and if they were questioned the model could be altered and the system could be revisited..

The other caveat was that Statistics South Africa was reliant on the information provided on the death notification. At this stage there was no obligation or clear guidelines on the medical practitioner to mention HIV/AIDS as the cause of death, as it was really a matter of their judgement as to the reason they provided. This was a significant constraint on the data produced.

Mr Kolweni asked whether legislation was not needed to expedite this process.

Mr Lehohla replied that one of the things that Statistics South Africa was looking at was that every doctor could use a computer programme to record the cause of death, which would then attach a coding mechanism which would be stored and processed. The main problem was the inefficiencies that were brought about my "paper pushing" and an electronic medium would be preferred. The technology was available.

Dr Gavin added that legislation on making HIV notifiable was clearly not Statistics South Africa domain. The other option would be to trying to speed up the collection of that kind of information and at the same time impose certain standards, such as capturing information immediately and digitally by providing specified choices for doctors to choose from. This would implement standards on data which would improve comparability. Statistics South Africa realised the importance of addressing the terms and definitions it used in its various data collection exercises, and it has considered the establishment of a task team to look at harmonising its definitions.

This would also lead to a rationalisation of the manner in which Statistics South Africa managed and storied its statistical information in terms of warehousing. It was realised that a critical element of being able to do more effective data mining and Statistics South Africa's own analysis of its data would be to have clearly defined standards and to standardise its information for improved comparability, which would enable people to undertake more meaningful analysis.

Mr Kolweni stated that the Statistics South Africa district authorities did not appear to have the necessary capacity, and asked whether the Financial and Fiscal Commission (FFC) was in a position to help the local authorities.

Mr Lehohla replied that the Department was in constant discussion with the FFC

Mr K Moloto (ANC) [NA] sought clarity on the areas of concern in relation to the determination of the GDP and consumer prices and the interim measures to deal with the shortcomings that have been identified by the Review Task Team (RTT) that was established to assess the quality of Statistics South Africa's outputs.

Me Lehohla responded that the RTT on economic statistics was established and was given impetus by the Consumer Price Index (CPI) debacle. A review was conducted and one of the issues that arose was the creation and generation of a strategy for economic statistics. There were strategies being implemented for correction of the CPI as well as for benchmarking the Gross Domestic Product (GDP). A team of experts was constituted and all three were Deputy Statisticians in their respective offices, and were thus sufficiently qualified to be heads of top class statistics agencies. They were part and parcel of the continually improvements.

Dr Ros Hirschowitz, Statistics South Africa DDG: Quality and Integration, added that with the help of international consultants the most important thing to realise about the reason for the break in series and why the rent component came up, was that Statistics South Africa used to conduct a rental survey. That survey was discontinued and Statistics South Africa believed it would be able to compile that information sufficiently from the household surveys. Unfortunately in household surveys which only had a 30 000 household sample and only 20% were paying rent, Statistics South Africa found that the data itself was insufficient. The aim was now to introduce quarterly household surveys that covered not only rents but also house purchases, house sales etc. because most countries, when they calculated CP's, calculated rental equivalents as well. Statistics South Africa has however only been taking rent and mortgage payments into account and it needed perhaps to include rental equivalents. This could lead to a further change in the CPI.

The rent survey would be piloted this year and it was hoped that it would be introduced in the next financial year. The CPI was always collected through postal questionnaires but Statistics South Africa now planned to go directly to outlets with well-trained people, and would be collecting prices directly. This was in accordance with the international practice. Statistics South Africa realised that the information it received from the postal questionnaires were not necessarily accurate enough, and the companies themselves were not being vigilant enough by themselves to enable Statistics South Africa to capture those figures. Statistics South Africa planned to send teams around the country to collect this information, and this was one of the reasons why Statistics South Africa needed to establish regional offices.

The other change in the CPI was with regard to the "basket of goods" and what it should consist of and the weights that should be given to each. This was normally revised every five years on the basis of an income and expenditure survey. Statistics South Africa was changing its method in this regard by collecting prices directly through a diary method. It was a much thicker questionnaire but it could be split into four different portions on food, transport, clothing and general purchases. This would provide a more accurate picture of the spending patterns as well as the weight that that should be attached to each parcel of spending.

Mr Moloto asked whether Statistics South Africa could provide an indication of the extent to which it undercounted the GDP.

Mr Lehohla replied that at the moment the quarterly review suggested that the trends were the same, and Statistics South Africa could not really pronounce as to whether there has been an under or overestimate of the GDP. These figures would only be available by November 2004.

Ms R Taljaard (DA) [NA] asked Statistics South Africa to explain the manner in which the respective agencies were working together in gathering this kind of data, because this did play a role in the reliability of the undercounting.

Mr Lehohla responded that the South African Revenue Service (SARS), the Department of Trade and Industry and Statistics South Africa "have taken a bite at a time". At the moment there was a bit of a problem because there were issues of dormancy with certain things that were registered, but Statistics South Africa did conduct an analysis on this. The problem remained the transfer of information across these three organisations. The ideal situation would be to have a single business number for all three, and this was in sight because they understood the urgency with which these needed to be achieved. Once this point was reached there would be better credibility in the updates.

Dr Herschowitz added that the slide in the presentation entitled "Snapshot of the South African business register" detailed the businesses considered. Statistics South Africa was thus trying as far as possible to arrive at a comprehensive and integrated business register. Government needed an integrated business register which provided government departments access to different components of information, such as turnover, business size and employment. No legislation existed at the moment which ensured that government departments had access to the information they needed from the integrated register.

She stated that Statistics South Africa worked closely with SARS and South African Reserve Bank (SARB), which calculated GDP from the point of view of consumption whereas Statistics South Africa was calculating it in terms of the value added through the production process and then through income. Statistics South Africa was then calculating it two different ways: the annual rebasing and the second was the short-term quarterly statistics that were being produced. Statistics South Africa did have discussions and did supply much information to SARB for them to calculate its figures from Statistics South Africa's expenditure figures.

These relationships were thus working but it would be a good idea, because of the vested interests of SARB to look at this working arrangement and perhaps have an independent agency that would compile all components of the GDP. This might be a very important step to take.

The Chair asked whether Dr Kirschowitz meant that this body would be independent of Statistics South Africa.

Dr Hirschowitz answered in the negative. She stated that she was referring to a body that was not involved in policy and in determining income inflation rates, or in managing the economy of the country as was currently done by SARB.

Mr Lehohla stated that this function must be performed by Statistics South Africa otherwise it would create a conflict of interest if SARB becomes involved. All the three surveys referred to should properly be conducted by Statistics South Africa. This capacity would have to be built up within Statistics South Africa over time.

The Chair agreed. She stated that the independent agency should actually be Statistics South Africa. This matter would be discussed with SARB when they appear before this Committee.

The Chair asked whether this meant that there was restricted access to this information, and asked who held the register.

Dr Herschowitz replied that the register was being put together by Statistics South Africa at the moment. All its information was collected under clauses of confidentiality, and it was thus unable to disclose that information to anyone else that it collected. The legislation was passed in 1998 to give Statistics South Africa access to some information from the VAT and Income Tax register. The problem was thus that if Statistics South Africa made the register generally available to other government departments etc. in an integrated way, there may need to be certain legislative changes to grant such access. Confidentiality affected any individual record and Statistics South Africa did apply algorithms to disguise the information.

Mr Lehohla added that the Statistics South Africa Act provided for confidentiality of information on listed groups of bodies and institutions. It also prohibits the use of any such confidential information in a court of law in order to prosecute on the basis of the information that was acquired.

The Chair asked whether these amendments would have to be effected to the Statistics South Africa Act.

Dr Herschowitz responded in the affirmative, and it might also be needed in other pieces of legislation as well

Ms R Joemat (ANC) [NA] stated that all the related government departments and institutions must work from the same book, because they all were collecting data but not that much reliance was placed on the statistics produced by Statistics South Africa. The departments and parastatals must be integrated to ensure that the same data collecting method was used.

Ms Taljaard asked whether, in keeping with the proposal to have the census every 10 years rather than every 5 years, a definitive decision has been taken to have it in 2011 and not 2006. The 10 year cycle created a problem with longitudinal data used for comparative purposes.

Mr Lehohla replied that Cabinet has taken a definitive decision that the next census would be held in 2011. He stated that Statistics South Africa could provide assurance on continuity in the data items, and could also provide an indication of the level of spatial resolution at which that particular data item would be applicable.

Mr B Mnguni (ANC) [NA] asked Statistics South Africa to explain why its figures for the conducting of the census has increased, when Mr Lehohla stated that the next census would only be conducted in 2011 and not in 2006.

Mr Lehohla responded that the large scale survey would absorb most of these funds, but Statistics South Africa would also like to deliberately strengthen two other aspects during this period. The first would be to conduct a large scale survey that would "quench the thirst" for a census. The second would be to build on the infrastructure so as to enable Statistics South Africa to become a means-directed leadership which looked at the population register and the dwelling register. The population register was compiled jointly with the Department of Home Affairs.

Ms Taljaard stated that there was clearly a need to continue to benchmark internationally when doing its revisions, and to adapt Statistics South Africa's methodology to those international benchmarks. On the other hand if revisions were conducted too frequently then the credibility would be questioned.

Mr Lehohla replied that the implementation of the Integrated Business Register (IBR) played a key role in this initiative, and the visible changes were a result of migrating to a more credible frame of reference. In this sense Statistics South Africa was between a rock and a hard place because change would certainly occur given what Statistics South Africa has migrated to, but how exactly those improvements would be managed to strengthen credibility was a major issue. This occurred with the backdrop of the CPIX debacle last year. The change that was currently occurring was not the same as the change that took place then, because the current change was anchored in a credible frame.

Dr Hirschowitz added that there has been a deliberate attempt in the field of economic statistics since 1996 to produce better economic statistics. There used to be the old Business Address Register (BAR) which was updated very haphazardly. Statistics South Africa did not have access to all the information it needed and it would have to use various sources to update the BAR. In 1996 Statistics South Africa realised that it had a pretty outdated register that was not updated in a very systematic way and in 1998 legislation was introduced to give Statistics South Africa access to information that it needed from the VAT and Income Tax registers. Since then Statistics South Africa has been working together with various government departments to try and put together an integrated business register that could be used by the country generally.

Since gaining access Statistics South Africa has worked very strongly on getting the integrated business register operational. At the moment its main source of drawing information and samples was from the VAT register, and Statistics South Africa maintained the register in which new businesses were added and the deceased businesses were taken off the register. In the next round of samples Statistics South Africa would be including the Income tax register, which Statistics South Africa did not include in its samples at the moment. It would list the non-VAT registered businesses but who were registered for income tax, and thus the formal business sector would be covered even better. Statistics South Africa believed that it now had a business register that was much better.

It took quite a while to draw the samples because, in September 2002 Statistics South Africa drew the first samples from the new business register, it was frozen and was called the business sampling stage and drew the samples. It took a long time to get classifications correct of businesses according to industry. The problem was that the VAT and Income Tax registers were legal registers, whereas Statistics South Africa needed a statistical register. One of the main tasks involved in putting together a statistical register was the classification of businesses according to industry, and the most time consuming component consisted of large businesses that performed more than one activity. They would then have to be classified into their various activities in order to assess what they contribute to each sector of the economy. The new business register was only at a usable stage in 2002.

Statistics South Africa had to measure the differences that were occurring between the old and the new statistics and, during the whole of 2003, Statistics South Africa ran parallel surveys. These indicated the extent to which Statistics South Africa was undercounting under the old BAR. Thus by moving to the new register and by running the parallel survey for a year, Statistics South Africa was able to understand the differences in levels. She stated that perhaps the one failing was that Statistics South Africa was so focused on getting the figures right that it did not give enough attention to the publicity and communication of the improvements made.

She stated that things have gone according to plan and the three consultants that Mr Lehohla mentioned have been very helpful in introducing these measures. These will be changed in future but they will not be as large and as dramatic as the present changes.

Ms Taljaard suggested that the National Treasury survey on administered pricing should also be included in this debate. To what extent were Statistics South Africa and Treasury working together on this if at all?

Mr Lehohla responded that a committee had been established by Statistics South Africa, Treasury, the South African Reserve Bank (SARB) and other participants to consider this. It was due to make a recommendation on how to proceed.

Dr Herschowitz added that Statistics South Africa was able to separate out and produce an administered price index without and without administered prices. It did have various meetings with SARB and Treasury to discuss the issue of administered prices. Some sort of an agreement was reached on what should be administered prices. There was a big difference between determined prices and administered prices. Examples of administered prices would include petrol and education, whereas an example of a determined price would be free health care in clinics. She stated that on 7 June Statistics South Africa would be presenting its findings on this to the Minister, but there may still be some definitional issues.

Mr S Asiya (ANC) [NA] requested Statistics South Africa to explain the measures Statistics South Africa has put in place to ensure it better controls its funds, as this would ensure better accountability.

Mr Lehohla responded that, when the decision was taken not to have a 2006 census, the Estimates of National Expenditure (ENE) had already been prepared. There was thus much work that needed to be done in order to provide the implications of the changing of the course of the ENE.

Ms Phuti Matlala, Statistics South Africa DDG: Organisation and Management, added that over the past 6-12 months the issues of spending over the previous years were revisit. Many lessons were learnt in areas in which Statistics South Africa could improve on the work it had already done. With regard to the ENE she stated that it had identified key projects per cluster or per DDG and, at the moment, Statistics South Africa was devising plans on the content and timeframes for these projects. All audit queries that arose from the past financial years have also been attended to.

Ms Taljaard requested that Statistics South Africa provide Members with documentation on its plans during the intervening period, the specific surveys that would be undertaken and the specific data that would be lost on the longitudinal side.

The Chair stated that this Committee could then have a briefing on the implications of not having the full population census but only the extended survey.

Ms J Fubbs (ANC) [NA] requested that this briefing include the expenditure trends which referred to the increase, as referred to earlier by Mr Mnguni.

Ms Mbuyazi (ANC) sought clarity on the extent of the interaction between Statistics South Africa and the provinces in compiling the poverty statistics

Mr Lehohla replied that Statistics South Africa had developed a unit that, in the first instance, provided the poverty measurement report in South Africa. This has since been institutionalised as unit that would continually work on poverty statistics, and works in collaboration with other government departments such as the Department of Social Development and in the main, Treasury and other research institutions. It aims to establish poverty lines and approaches to poverty, and there would be much dialogue on this

Ms Mbuyazi asked what other means could be used to gather the information Statistics South Africa produces, as those in the deep rural areas did not have infrastructure such as computers or access to the Internet.

Dr Hirschowitz responded that the income and expenditure survey was used as the main tool for measuring monetary poverty. It was not the only measure of poverty but it did provide the expenditure figures which was often more accurate than income. Statistics South Africa has been developing poverty lines on the basis of expenditure and it plans to take those measures across through a series of regression analyses into the census, as was done in 1996. This would enable Statistics South Africa to indicate an overall average expenditure for each remuneration area in the country, but would still have to be planned.

She stated that Statistics South Africa did compile data on purchasing power parity poverty in terms of $1 or $2 a day earning, except that this would be calculated on the grounds of spending. A study was also conducted on people according to their relative poverty, who would fall into the 20th percentile, poverty be province as well as poverty in terms of access to facilities. As part of the calculation of national accounts, Statistics South Africa also conducted a social accounting matrix and published the findings.

Mr Mnguni asked Statistics South Africa to explain the National Statistics System (NSS) and the extent to which both national and provincial government departments were using the system.

Secondly, Mr Mnguni asked Statistics South Africa to explain the extent to which it had persuaded policy-makers to go beyond a mere acceptance of an NSS in principle and to go forward and implement

Mr Lehohla replied that in 2002 Cabinet approved a NSS. Much work has been done on this and was primarily focused on the raising of consciousness on the role and need for co-ordination. This aspect was completed. Statistics South Africa was working on a series of indicators for cities and the Limpopo Province was being used to test how best it could use its information management systems as well as the census information in implementing the NSS. More importantly the issue around standards around standards and definitions was taking priority in the organisation.

Another instrument that makes NSS a living organisation was in the manner in which positions were advocated, because the statistics must be viewed on the same basis. Co-ordination mechanisms were also being explored, as mentioned in the President's State of the Nation Address. Much work was thus being done with regard to the NSS.

He stated that Statistics South Africa would be implementing it in the Education, Local Government and Home Affairs departments in 2004, and if rapid progress were made on these then more would be taken on.

Mr E Sogoni (ANC) [Gauteng] asked Statistics South Africa to explain whether the budget allocated for the communications directorate was for internal communication or external communication, because it was a large amount.

Mr Lehohla responded that it included the budget of the Office of the Statistician General as well as those of the Deputy Directors General, and it was not only communications. Efforts were being made to strengthen external communications as well, because Statistics South Africa was very well aware of the fact that it concentrated so much on perfecting the product but had no clue as to how to take it to the market in a form that the users themselves would appreciate.

Dr Hirschowitz added that Statistics South Africa did have publications that reached its users, it did distribute some of them to schools and it did distribute very widely at the time of the census. It did have user groups that were used when Statistics South Africa developed a questionnaire that was used to measure aspects of a specific issue. These would involve national departments.

Mr Kubega (ANC) asked whether Statistics South Africa had devised a specific programme to ensure that schoolchildren become aware of pursuing a career in statistics.

Mr Lehohla responded that Statistics South Africa addressed this issue with much vigour. In 1996 the census was launched by the Ministers of Education and Finance at three schools here in Cape Town, which was the beginning of providing inspiration in school children for the study of mathematics and statistics. At the end of that process Statistics South Africa had collected information on conditions of school as well as the careers followed by school children. More recently Statistics South Africa distributed 300 computers across the country, including the information that arose out of the census at schools, and those particular schools could then analyse their own data about their school.

With regard to the curriculum Statistics South Africa was engaged in discussions with the Department of Education to incorporate the results of the census at schools into the curriculum so that people could begin implementing that.

Mr Y Bhamjee (ANC) [NA] asked Statistics South Africa to indicate whether equity to effect transformation in the statistics field was an option.

Mr Lehohla responded that Statistics South Africa boldly stated that it would provide training and stated that that training programme was continuing. It began with 21 candidates in 2001 and all have since been fully integrated into organisations. A second group of 12 candidates where then put on the training course and they were expected to return this year. A third group left for training in Tanzania in this year. At the same time there were 9 candidates being trained at Wits University as well as 10 persons who were studying advanced methods in statistics. South African universities did not provide the kind of training that a statistical agency ideally needed, as the training needed to be adapted to the needs of statistical agencies. Statistics South Africa was implementing a virtual institute of statistics with advanced courses on statistics estimation, sampling etc. on offer.

This resulted in the management structure of the organisation becoming more representitive. In the causes of death project referred to earlier by Dr Gavin, Statistics South Africa ensured that it was performed by people with disabilities, who also conducted the processing. A number of them secured jobs after the completion of that task. In fact they performed better than those without disabilities. There was thus a business-orientated case for differentiating that kind of work.

Dr Hirschowitz added that there were different levels of training that Statistics South Africa was trying to put in place. The first was the schools and the census at schools as described by Mr Lehohla, and how that was taken further with the Department of Education in terms of curriculum development. Courses were also introduced for journalists to try and educate them on statistics, and a very successful programme was conducted with the SABC in which certain concepts were discussed with them over a three-day period. It was hoped that this training would be expended to other institutions and government departments.

Statistics South Africa provided training for its own employees. Efforts were being made to have all Statistics South Africa's training courses registered with the South African Qualifications Authority for inclusion as part of the qualification structure in the field of statistics.

Mr Bhamjee stated that the presentation indicated that there was a great shortage of skilled people and qualified staff within Statistics South Africa and requested Statistics South Africa to indicate how exactly it would address this. Further, what measures were being taken by Statistics South Africa to ensure that those people that possess the necessary skills were taken up by the industry. He also asked for an indication of the gap between the African, Indian, coloured and white employees in Statistics South Africa.

Ms Joemat requested clarity on the in-house training that was provided which enabled people to get onto their career paths.

Mr Lehohla responded that the precise figures on the in-house training breakdown as well as the staff establishment broken down on a race and gender basis could be supplied to Members at a later stage.

Ms Fubbs suggested that the measures, indicators and targets employed by Statistics South Africa be refined in order to reflect the quality of the statistics produced. Furthermore, she asked Statistics South Africa to explain the figures referred to under "Public sector national government", "public sector provincial government" and "private sector manufacturing" in the slide entitled "Customer satisfaction index 2002".

Mr Lehohla responded that this was always a major challenge, as quantitative indicators were easier to arrive at then qualitative indicators. A good number of the 22 indicators established for the ten year review were qualitative, and they became difficult to handle. Statistics South Africa would have to work on this in arriving at a balanced scorecard.

The slide referred to indicated the number of provincial departments that responded, and they might not all be finance departments. The manufacturing figure indicated manufacturing in general, but a more refined model could be worked on. He stated that he would ask his colleagues to compile a report on the information requested by Ms Fubbs, short of breaching any confidentiality constraints.

Ms Fubbs asked Statistics South Africa to explain its measures to track and monitor all its programmes.

Mr Lehohla replied that Statistics South Africa has implemented a management information system that firstly identifies from the strategic plan those items that Statistics South Africa said it would implement and, it Statistics South Africa progressed, the tracking system would monitor the implementation. At the moment 80% of the work would now come through the management information system.

Mr Y Wang (ID) [NA] asked Statistics South Africa to explain its motivation in sending staff members to very specialised courses when providing in-house training, as two of the courses mentioned were for IT and computer programming skills. Surely they should be taking statistics courses.

Mr Lehohla responded that the nature of statistics and IT was blurred, because statistics was part and parcel of IT. He stated that IT was thus absolutely critical and strategic to statistics operations. In fact it was so important that it mapped to the fourth quadrant to the point that the two could not be separated.

Mr Wang asked whether the data mining techniques were used in conjunction with the data warehouse, as this would unlock the full potential. It would also lead to integration of the information to other government departments such as the Department of Home Affairs. He asked whether a task team was established to work with the Department of Home Affairs and Department of Health on this integrated database.

Mr Lehohla replied that Statistics South Africa had a number of students at Wits University who were studying data mining and, as a confidence of this exercise, Statistics South Africa had confidence in the population register. He stated that Statistics South Africa and the Department of Home Affairs were working jointly on the turnaround strategy on Home Affairs National Identification System (HANIS) in order to create the three registers: the dwelling address, the population register, the business register. He stated that Statistical Analysis Software (SAS) was used which contained the data miner, and one licence cost R1,5m.

Ms Fubbs stated that the presentation indicated that adequate funds were not available for the income and expenditure survey and the dwelling survey, and asked Statistics South Africa to explain the measures put in place to address this.

Mr Lehohla responded that this was also a planning and integration across government issue as he believed that if Statistics South Africa rationalised well it would be able to deal with the income and expenditure survey.

Dr Hirschowitz added that Statistics South Africa would use the additional funding it would receive for census 2006 to run the extended household survey, the address register system and to improve its geography and its geo-positioning of every dwelling unit in the country. These funds would then enable Statistics South Africa to improve the income and expenditure survey. It was budgeted for but the diary method was much more expensive, and Statistics South Africa was requested additional funds for the next financial year.

Ms Fubbs asked Statistics South Africa to explain its plans to increase the accessibility of civil society to its statistical information.

Mr Lehohla replied that if Statistics South Africa were to work through the Multi-purpose Community Centres (MPCCs) this could be alleviated, but much more training was needed. The communication of complex statistical information in a format that was readable and accessible to such organisations was the main constraint, and it was thus not a technological issue.

Concluding remarks by Chairperson
The Chair thanked Mr Lehohla and his team for the interaction, and the Committee looked forward to future engagements with Statistics South Africa in fulfilling its oversight function.

The meeting was adjourned.


No related


No related documents


  • We don't have attendance info for this committee meeting

Download as PDF

You can download this page as a PDF using your browser's print functionality. Click on the "Print" button below and select the "PDF" option under destinations/printers.

See detailed instructions for your browser here.

Share this page: