1. GENERAL CONTEXT
Previous  1.2. Analysis of the demand for NSS support  Next


1.2.1  Process to define a statistical demand

Box 1 :  Transforming basic data into aggregates and other economic and social indicators

Basic data
Data recorded by the person directly concerned,
               in accounts for a company,
               in his/her memory for an individual,
               in a register for an administration,
based on the concepts of the user rather than the statistician’s.
Collected data
Those are stand-alone anonymous data since they have undergone consistency checks, amendments and sometimes expunction. Such adjustments are essential to correct errors and specifically such that would cause erroneous interpretation of questions by the respondent.
Data
These are results of surveys or processing of data collected from administrative documents, collected or aggregated data constituting an average statistical unit (average individual) of a sub-population. The level of classification of individuals into sub-populations depends on the size of the sample, homogeneity of the population and frequency of the variable under observation.
Aggregated data
Data calculated on a model that uses and processes all available input and data sent in by several data collection and processing systems, including different statistical entities. By way of example, data from national accounts, population estimates, unemployment projections, employment figures, etc.
In general, these data are highly aggregated. They partially correct bias and errors in statistics. Such aggregated data are, to a certain extent, projections that are hurriedly compiled but which must be revised when new data becomes available.

For the purpose of providing efficient support to NSS of its Member States and to the regional standardization process, AFRISTAT and Member States must have a common vision of problems plaguing NSS and attendant solutions.

A statistical system is a productive sytem and must be distinguished from a statistical information system. NSO and their constituent sectoral statistical services, jointly and severally, produce statistical information and data which actually become relevant when they are organized in a statistical information system. This must not cause confusion between the producer and the product. The statistical system has an ultimate objective: to meet demands which can be funded for statistics.

Box 2 :  Demand which can be bankable

Statistical demands
As a productive system, the statistical system must satisfy the needs of users through regular and sustainable production.
A demand is considered to be bankable where production expenses are fully covered in order to handle the user’s needs. Such a demand applies only to needs that have received funding.
Statistics are essentially public property made available to all citizens. Its production should be financed to a large extent, by public funds depending on the production capacities of NSO and sectoral services. The production of statistics is therefore based on a tripartite agreement involving financiers, notably public authorities, users (generally driven by international initiatives) and producers.

Statistics is a tool used in guiding decision-making especially where they substantiate an evaluation, a study, policy impact simulations, etc. Hence, the appropriate services and bodies responsible for performing such tasks for administrations, large enterprises, employers’ organisations, labour confederations, civil society, etc., which are the main users of data and statistical reports thereby influence the demand for such. However, they only influence demand to the extent that they too have to cope with strong requirement from decision-makers, notably policymakers and corporate executives. Accordingly, the development of statistics is driven by the development of decision-making guidance services such as planning, forecasting, development strategy and poverty reduction.

There are four major stages in the production of data: (i) designing and updating sample frames and directories, (ii) data collection and processing, (iii) designing statistical reports and making relevant analyses, and (iv) disseminating data and statistical reports.

Directories and sample frames are indispensable tools to statistical services. They index villages and neighbourhoods, enterprises and companies, government services and various educational instutions as well as health centres, etc. While designing a directory may require start-up funds that are seldom high, its maintenance is indispensable and expensive.

The collection and processing of statistical data is the core activity of every statistical service. Data may be collected by way of a survey or consultation of administrative reports. In both cases, downstream the collection, the processing chain is similar. Designing a collection and processing system is a huge investment. Mass and regular production alone can guarantee a return on the investment1 . It is dependent on the availability of exhaustive and updated directories or sample frames as well as sustainable collection instruments (network of experienced and qualified controllers and surveyors available countrywide) used in surveys.

Box 3 :  The mandatory trade-off between the three charateristics of statistics to be produced from a survey

Depth of description
The more a survey seeks to measure different variables, the more the questionnaire is complex and the more time it takes to source the information, which is pretty resource demanding and especially, may distort the survey and downplay the quality of the expected responses.
Accuracy of data
This refers to the number of sub-populations from which data is sought. The higher the number, the larger the size of the sample which will entail further sampling and higer costs for the survey.
Frequency and timeliness of production
While a repeated survey is less expensive than a one-off survey, repeated surveys have a price. Moreover, frequent data production calls for immediate data use.

Lastly, statistical reporting is a key step involving multidisciplinary teams that aims at understanding socio-economic trends by analyzing collected data that becomes relevant only through the widest dissemination of its findings.

1.2.2  Specific problems facing NSS of AFRISTAT Member States

National statistical systems of AFRISTAT Member States encounter difficulties that can be summarized under three points: (i) bankable data demands are not clearly defined, (ii) NSS financing is inadequate and (iii) their institutional capacities unsuitable to handle growing and diverse demands. In short, insufficient statistical culture and declining permanent consultative bodies such as the National Statistical Council are the main drawbacks to statistical development in these countries.

During structural adjustment programmes, statistical production dwindled for the following reasons: Ministries of planning saw their powers curtailed while NSO, when they could, produced macroeconomic statistics with a lot of difficulties and often behind schedule. Today, there is seemingly a sustained demand for data on the monitoring and evaluation of poverty reduction strategies and MDG.

In the absence of pressure from national users, policymakers have seldom allocated the requisite resources to national statistical systems to produce statistics and less so, to sustain its production capacity. There is an unfortunate tendency to believe that external donors should fund statistical production since they alone deem it important. This further weakens NSS.

In many countries, statistical capacities are not maintained and have become so weak that national statistical systems are unable to meet unexpected demand. Several years of sustained efforts are necessary to revamp the quality of directories and sampling frames or to rebuild teams of competent professionals.

1 In fact data users are generally interested in the trends rather than the variables.


Previous  Upper level  Next
1.1. Overview of the statistical environment in the AFRISTAT intervention area prior to the launching of ASPA Table of contents 1.3. Achievements, strengths and weaknesses of AFRISTAT