Strategy for the Implementation of
the Addis Ababa Plan of Action for
Statistical Development in Africa in the 1990s
Strategy for the Implementation of
the Addis Ababa Plan of Action for |
I. INTRODUCTION
The working group meeting on the implementation of the Addis Ababa Plan of Action for Statistical Development in Africa in the 1990s, held in Nairobi in July 1991, further reviewed and elaborated on the principles, objectives and recommendations of the Plan of Action. It formulated a document entitled "A Strategy for the Implementation of the Addis Ababa Plan of Action for Statistical Development in Africa in the 1990s". This strategy document was subsequently considered and approved in March 1992 by the seventh session of the Joint Conference of African Planners, Statisticians and Demographers and adopted in April 1992 by the Twenty-seventh session of the Commission/eighteenth meeting of African Ministers responsible for Economic Planning and Development.
The Strategy recognized that measures for enhancing the capacities of African statistical services will have to be country specific, taking into account the different stages of statistical development. However, it is also acknowledged that there are general guidelines that can be formulated to assist countries in drawing up and implementing plans to enhance their statistical capacity. The Strategy therefore recommends that each country of the region should undertake Needs Assessment and Strategy Development (NASD) exercises as the first step towards setting up efficient and sustainable statistical systems.
The recommendation emanated from the realization that the level and quality of statistics had deteriorated in most African countries towards the end of the 1970s and throughout the 1980s. This was due, among other reasons, to economic crises that led to cuts in public expenditures, low priority attached to quantitative information in development planning, low status of statistical services vis-a-vis other government services, poor management of statistical services and competition between external and internal demands for statistical data.
In this connection, it must be stressed that there is the need, even in countries with more developed statistical systems, to re-assess and re-evaluate the adequacy of the statistical system in order to ensure appropriate and efficient service. This is even more necessary under the prevailing situation of the statistical system in most African countries. The NASD is, therefore, considered a prerequisite for the formulation of a national statistical development plan.
The exercise is the responsibility of individual countries. The relevant initiative may be "demand based" as when promoted by users. This is the situation where NASD is conducted, for example, in response to some expression of discontent with the available statistical service. Alternatively, it may be "supply-based" in which case it is carried out on the initiative of the statistical agency due to its concern to improve the appropriateness and efficiency of its service.
II. THE PROPOSED GUIDELINES FOR NASD
The present guidelines are intended to assist African countries in carrying out the NASD exercises. They are based on recommendations in the strategy document. Also, some ideas have been drawn from the United Nations Population Fund (UNFPA) "Manual for Needs Assessment and Programme Review and Strategy Development" as well as from reports on needs assessment missions undertaken by the British Overseas Development Administration (ODA) in Africa. They are not to be regarded as inflexible list of actions but rather a framework of alternative procedures that countries may consider in undertaking NASD exercise.
The tasks involved in NASD include programme evaluation, analysis of user needs and priorities, etc. They are presented under the following headings:
• Organization of NASD
(a) objectives
(b) responsibility for its organization
(c) terms of reference for the exercise
(d) composition of the team and criteria for selection of team members
(e) timing, duration and frequency
(f) background documentation and consultations
(g) support facilities
• Output(s) of the NASD
(a) content of the report
(b) draft statistical development plan
• Follow-up actions
(a) presentation of report to appropriate bodies
(b) implementation of recommendations
There is the need to ensure continuous improvement and effectiveness of the guidelines. Comments from interested parties are, therefore, encouraged and welcomed. Indeed, these guidelines are regarded as preliminary and are to be reviewed in due course on the basis of lessons learnt and comments received.
(a) Objectives of the NASD
The main objective of the NASD is to "find out what the current and future data needs of the primary users of statistics are likely to be" and how the national statistical system should react to satisfy these needs within resource constraints.
In assessing the current and future data needs as well as requirements of the statistical system, it will be necessary to evaluate past use of data, determine effects of lack of data, if any, on planning, policy making and other work and also ascertain the national statistical capacities. These will provide a basis to propose any required changes in the statistical infrastructure. In summary, the objectives of the NASD are to ascertain:
(a) What data have to date been made available to users;
(b) What data have been used in the past;
(c) What are the data gaps;
(d) How data gaps have constrained planning, policy making and other work;
(e) What are the current and future data needs of the primary users, and what priorities are to be attached to these, and to what extent do users value timeliness, completeness of coverage, accuracy and other aspects of statistical quality relative to the cost of providing such quality;
(f) What will be the statistical programme to satisfy these requirements and what changes will be required to the existing programme (eg cancellation of certain collections, revisions to others, improved use of existing sources, introduction of new collections, etc);
(g) What institutional and legislative reforms are required to improve the effectiveness of the Statistical System.
The NASD will generate at least two outputs. The first should be the review of current and future needs, the last should be a long-term strategy document for strengthening national capacity for prioritizing and eventually meeting those demands. The strategy should include a timetable and budget and should provide a reference document to be put to donors for financial and technical support.
(b) Responsibility for organizing the NASD
The NASD will have more authority if the NASD team is established by a senior government official, such as the President himself or the Minister who is responsible for the national statistical agency in each country.
In some countries, the team might be appointed by the Permanent Secretary of the responsible ministry or even by the director of the national statistical agency. But if possible, a more senior official would be preferred. The designated official would appoint the Needs Assessment and Strategy Development team, and issue the terms of reference.
The state of statistical development in the country should guide the practical details of setting up the NASD. Each country will have to decide for itself the extent to which it will require external assistance.
As a first step, directors of national statistical agencies may wish to submit a memorandum to their Minister on the issue of needs assessment and strategy development. The Minister would proceed to set up the NASD team. The UNECA Statistics Division, in its capacity as secretariat of the Coordinating Committee on African Statistical Development (CASD), should be informed about the setting up of such teams and, where applicable, the external agencies which have been invited to participate. This would enable to facilitate the monitoring of the implementation of the Addis Ababa Plan of Action for Statistical Development in Africa in the 1990s.
(c) Terms of reference of the NASD
The NASD team is expected to perform the following tasks:
A. Prepare a set of recommendations for statistical development on the basis of past experience by:
(a) Ascertaining country policies and priorities in respect of economic and social development, population planning, migration, economic activity and conditions of employment, status of women and children; environment, etc;
(b) Identifying the availability and timeliness of existing statistical outputs, and the extent to which they are understood and used;
(c) Assessing the adequacy of existing statistical systems design in collecting the statistics which are required in the most efficient and cost-effective manner. This might include an assessment of whether there are too many questionnaires, whether there is duplication within or between questionnaires, whether the questionnaires are readily understood and easily completed, whether more than one national agency is collecting similar data (and the extent of collaboration between different statistical producers), whether there is a coordinated and integrated central business register, etc.
(d) Assessing the extent to which existing statistics have satisfied (and are expected to satisfy) users in setting and monitoring national plans and policies, in providing macro and sectoral indicators, and so on;
(e) Reviewing the extent and frequency of collaboration and consultation between data producers and users (and between different data producers in the same field);
(f) Ascertaining past and expected future gaps in the available statistics, assessing the constraints that these gaps have placed on the work of users, assessing the priority given by users to issues of data quality (completeness, accuracy, etc) versus timeliness and cost, setting priorities for existing or required statistical outputs;
(g) Undertaking an inventory of national commitments to provide statistics to international agencies, and assessing the costs and suitability of these commitments. Making recommendations to cease or revise these commitments as necessary, or to request international review of external agency requests;
(h) Reviewing the statistical programme and recommending at least a minimal set (or preferably an optimal set) of new or revised statistical outputs and the discontinuation of low priority and unused data. In particular, the team should recommend a detailed programme covering an immediate planning period (such as three years) and give a broad perspective of a long term programme (covering a longer period) - see (B) below;
(i) Assessing the infrastructures and institutional capabilities relating to project management, data collection, data processing, dissemination and statistical analysis, especially in respect of the national statistical system but also covering analytical capacities in the areas responsible for planning and policy making;
(j) Assessing the financial, material and human resource constraints to statistical development;
(k) Making recommendations concerning statistical advisory or policy making bodies, statistical legislation, organization and management of statistical organizations and services;
(l) Reviewing and making recommendations concerning statistical training and research in the country, including "in-house" training, on-the job training as well as more formal training by government and non governmental bodies;
(m) Reviewing previous external assistance to the statistical development of the country, covering issues such as the extent and sources of funds, the extent and content of assistance, the degree of capability building and its sustainability, the coordination of assistance, etc, and making recommendations on these matters to guide and make the most effective use of possible future assistance. The team should also identify and cost any external aid inputs which are envisaged to be achieved in the statistical programmes which the team recommends. This review may be linked with each country's NATCAP/ exercise;
(n) Commenting on likely or possible constraints to the implementation of programmes for statistical development in the country.
In undertaking (a) to (h) above, a sectoral approach might be useful by breaking the problem into issues affecting economic statistics, demographic and social statistics, and other statistics (including those relating to economic activity, agriculture, environment and natural resources).
B. Formulate a draft national statistical development plan covering a period of about 5-10 years on the basis of its findings and recommendations by:
(a) defining clearly the objectives of the plan, the problems to be addressed during the plan period and possible solutions;
(b) forecasting human resource requirements by type of experience and qualification, and proposing national training goals to build the necessary human capacity;
(c) preparing detailed costings for the development of the institutional capacity and delivery of the statistical programme;
(d) relating the proposed national statistical development plan to the Government's strategies, priorities and planning cycle.
The draft national statistical development plan would need to be realistic and practical in the light of national capacities and development. It should not be overly optimistic, and as far as possible, should be able to be implemented within national resources. Access to external financing should, as far as possible, be limited to filling funding gaps.
(d) Composition of the team and criteria for selection of team members
As stated in the "Strategy" document, "no single model" for the composition of the NASD can be proposed. The size of the team is variable. It is recommended that the team be chaired by an eminent personality who should preferably be from the user side (an economist, economic geographer, sociologist, statistician or other social scientist) and be able to approach the Minister responsible for statistics directly if the need arises. There should be some balance between statisticians and users and their specialization should cover all major fields." Ideally, the members should all be senior and respected nationals drawn from government, private or parastatal (including university) sectors.
As already mentioned, some countries may require external assistance for the NASD and may wish to invite appropriate external agencies to participate. It is probable that these agencies will have had (or be expected to have) considerable experience in supporting the statistical programmes in the country. They may be UN agencies, other multi-lateral or bilateral agencies as the country deems appropriate. It is assumed that each agency will be responsible for all costs associated with their representation in the team.
Great care will need to be given to the choice of the best possible persons for the NASD. The selected persons are likely to have the following qualifications:
(a) appropriate knowledge of quantitative analysis and/or data collection;
(b) capacity for analytical and critical thinking;
(c) able to express themselves clearly and concisely in writing;
(d) able to work well in a team and under time pressure;
(e) available for the period of the review; and
(f) previous experience in evaluation of programmes/projects.
For any external advisers, these qualifications should be extended to include:
(g) previous working experience in developing countries; and
(h) appropriate language qualifications.
External advisers will be fully funded by their agencies. These agencies may wish to ensure that provision is made in their budgets for consultancies, travel funds, etc for this purpose.
(e) Timing, duration and frequency of the NASD
The duration of the exercise will depend on the circumstances and the stage of statistical development in each country. Some countries might require only eight weeks, while others might need up to six months. As a general principle, enough time should be allowed for all main users and institutions associated with the statistical system to fully understand and participate in the assessment process.
It will take some time to set up a team comprising such important and respected members, each of whom may be available at different times and these need to be reconciled. Also, because of their likely limited availability, the NASD will probably have only limited duration.
Four parts may be considered for the NASD exercise: prior preparation by team members, a joint period of assessment and deliberation arriving at recommendations, a period of report preparation to present the findings of the NASD and a period of preparation of the draft national statistical development plan. It may not be necessary for all team members to be present for the entire duration of the exercise and the team leader may be able to designate responsibility for particular sectors, etc to individual members of the team. However, there should be periods, especially when the team is deliberating its conclusions, when all team members should be present.
There is much to be done by the team. The NASD will have a period of individual literature search and background reading before meeting as a team for the assessment and deliberation period. These tasks are so important to the effective statistical development of the country, that they cannot be rushed.
It is also acknowledged that NASD should not be a "once off" exercise in each country, but rather an iterative periodic process which is also likely to be an integral part of the process of national development.
Statistical development should not be done in isolation, but must be compatible with the development objectives of other sectors. There would be little value in proposing a major expansion to the work force in statistical activities, if current government policy is to reduce civil service wage bills. Statistics will also be required for identifying problem areas requiring development intervention, and for monitoring and evaluating national development.
The existence of a national development plan will clearly provide a valuable source of sectoral priorities and needs, and as such a plan is reviewed (perhaps quinquennially) so too should the national statistical development plan. Even in countries where there is no national development plan, changes in sectoral priorities will require changes in the national statistical development plan.
(f) Background documentation and consultations
The team members should read as much background material as possible. A background document or briefing paper should be produced by the director of the statistical agency for the team which will identify and provide an analysis of the issues which the team should address. This paper should present facts without judgement and might be presented using similar headings to those recommended for the final report of the NASD. It should cover as much as possible topics such as existing policies and future plans; legislation; institutional infrastructure and
arrangements; human and technical resources, equipment, supplies and training; planning and management resources; local and external funding; details of previous evaluations (if any); etc.
In addition, the NASD team should consult documents such as National Development Plans, previous project evaluations, previous assessments of statistical capacity, all statistical publications, and other quantitative outputs of government, universities and other parastatal and private organizations.
The NASD will also benefit from reading the "Strategy" document and from a UN Statistical Division technical study on the structure and organization of national statistical offices undertaken in the mid-1980's.
The NASD should have discussions with all major producers and users (or potential producers and users) of statistics in the country. Appointments for these discussions should as far as possible be set up in advance of the exercise (see next section).
The users include the Ministries of Planning and Finance and the sector Ministries, the Central Bank, academic and research institutions as well as the private sector. Representatives of important sectoral groups such as employers, employees, agriculture, trade and so on should also be consulted.
In addition, the NASD should consult with statistical producers to ascertain their priorities, future plans, resource constraints, degree of integration, standardization of definitions, concepts and classifications, training arrangements, and other matters relevant to the terms of reference of the team. The principal producer would presumably be the government statistical agency, but other possible producers will be the statistical units of ministries (which may or may not be outposts of the central statistical office), the central bank, universities, selected non-government organizations and so on.
(g) Support facilities
The NASD will need local support prior to and during its operation. Preparations for the NASD include preparing terms of reference for the team, choosing the team members, sending invitation letters to them and undertaking other administrative aspects of their recruitment, consulting with the team leader to determine which appointments are required and then setting up these appointments; preparing a briefing paper for the team, assembling sets of relevant documentation; etc.
Wether external assistance is required or not local support for the NASD operational in terms of administrative and secretarial support, office accommodation, transport, stationery supplies, and other administrative and logistical support, etc. should be provided by the ministry responsible for statistics and or the national statistical agency.
III. OUTPUTS OF THE NASD
It is not necessary to be too specific on whether there should be one output or several. If the NASD is implemented in a phased manner, then an output might be produced at the completion of each phase.
Nevertheless, two main areas should be covered by the output(s), namely (i) recommendations arising from the statistical needs assessment and strategy development and (ii) a draft action plan for national statistical development (including a budget) based upon those recommendations.
(a) Content of the report
The output(s) relating to the statistical needs assessment should state current and previous features as well as likely future changes in these, and should cover the following issues:
(a) Country background and setting, including geographic, social, demographic, economic and administrative features.
(b) Country policies and priorities, including economic policy, population policy, development objectives, National Development Plans, sectoral development plans, etc. Government policy in terms of government staffing, recruitment, training and remuneration may also be covered if this is considered relevant to statistical development in the country.
(c) Infrastructures, institutional capacities and objectives in the areas of planning, policy making and statistics. Managerial and administrative abilities and capacities should also be evaluated in this section. The section should also cover the structure of the statistical organization, including where it fits into the government ministry structure what other agencies or councils are related to it, what is the sectional structure of the national statistical agency itself, and whether any statistical functions of the agency are decentralized or not (See also paragraph 34 above concerning the UNSTAT study on this topic. A later more elaborated version of these "Guidelines" might include some suggested alternative structures as annexes). This section might also consider the security of tenure of the director of the national statistical agency, and cover issues such as fixed term contracts in order to avoid high turnover in executive management, and annual performance evaluations of management
(d) Data availabilities, including the extent of data analysis, dissemination and utilization, and mechanisms and procedures for data collection, processing and analysis. Data gaps are to be identified as well as constraints arising from these gaps.
(e) Statistical research and training aspects, including infrastructure, courses offered, etc. This section will cover both in-service training and formal training at institutions.
(f) External assistance, including whether multilateral, bilateral or non-governmental assistance the source and extent of funding, and the coordination of this assistance.
(g) Recommendations arising from the above evaluations, including recommendations on the future statistical work plan, statistical priorities, legislation, methods for ensuring a smooth and regular communication between users and producers of statistics, the formation of a statistical advisory board (or statistical policy council)/, organizational structures, etc.
The output(s) may also contain annexes on budgets, staffing, organization and structure charts lists of statistical outputs, etc. Annex I to this paper presents the above recommendations in the form of a table of contents in order that the reports for different countries might have a standard presentation.
(b) Draft statistical development plan
The draft national plan for statistical development could include such issues as overview of statistical development; major challenges for the 1990s within the framework of the Government's overall development objectives; objectives, strategies and priorities of the plan; statistical information system, including among others organization and management of statistical services; statistical committees and statistical legislation; statistical priorities and development of human resources.
It is not intended at this stage to propose a table of contents for the draft national statistical development plan as the situation and conditions vary from one country to another.
IV. FOLLOW-UP ACTIONS
(a) Presentation of report to appropriate bodies
The team leader will be responsible for the presentation of the final report and draft plan. The NASD report and the draft plan for statistical development should be addressed to the designated official (eg Minister responsible for statistics) who appointed the team, and copied to the relevant permanent secretary and the director of the statistical agency. In some countries, the output(s) of the NASD might be submitted to a higher or lower level as appropriate.
The draft plan for statistical development should then be circulated among users and producers for their comments and suggestions which should be taken into account in the revision of the draft by the body overseeing the work of national statistical service.
(b) Implementation of recommendations
The Minister responsible for statistics (or another designated senior official),after due consideration and approval of the NASD report and the revised draft plan for statistical development, may take the necessary action to obtain the Government decision on statistical development. The Minister may then direct the permanent secretary and/or the director of the statistical agency (and other Government officials as appropriate) to implement the recommendations and the national statistical development plan after acceptance by the Government.
The implementation may entail the preparation of revised statistical legislation; the setting up of new statistical advisory or policy boards (see also footnote 2) and new mechanisms for improved consultations with users; preparation of a budget document showing details of inputs and final outputs for each financial year; formulation of specific project(s) for external assistance, including implementation strategies, technical specifications, cost estimates and coordination of assistance; etc.
It should be pointed out that some countries have arranged for major statistical development projects (for example, the implementation of a national household survey programme or population census) to be included in their overall national development plan in order to attract government financing (and possible external financing). Such an action could also be taken in respect of the total national statistical development programme in order to ensure that statistical development is given funding and recognition as a topic of national priority
Finally, the implementation of the recommendations and national statistical development plan needs to be followed up, monitored and periodically reviewed and updated. This could be one of the functions of the suggested Statistical Advisory (or Policy) Council, or one of its sub-committees.
V. CONCLUSION
As already indicated the present guidelines have been prepared in order to assist African countries in the implementation of a NASD. In view of the different stages of statistical development now experienced by different countries, it is difficult to generalise. Countries may wish to adapt these various suggestions to suit the appropriate conditions in their country.
Annex I
POSSIBLE CONTENTS OF NASD REPORT
A. Country background and setting (past and present
features, as well as likely future changes)
1. Geographic features
2. Social features
3. Demographic features
4. Economic features
5. Administrative features
B. Country policies and priorities (current situation
and likely future changes)
1. Economic policy
2. Population and social policy
3. Development objectives, National Development Plans
4. Public service human resource management policy
C. Infrastructures, institutional capacities and objectives, including managerial and administrative abilities and capacities (current situation and likely future changes)
1. Planning
2. Policy making
3. Statistics
D. Statistical system and user issues
1. Degree of communication between users and producers of statistics
2. User priorities on data quality (accuracy, completeness of coverage, etc) relative to timeliness and cost
3. Data availabilities (taking into account all major production of statistics in the country)
4. Data gaps and constraints arising from these
5. Data duplications or inconsistencies, and extent of collaboration between different data
producers (if any)
6. Data collection methodologies (extent of data integration, degree of harmonization of questionnaires, problems in timeliness and quality of response in data collection)
7. Data processing methodologies (planning, timeliness, standardization of software/hardware. etc)
8. Data dissemination methods and coverage
9. Methods and extent of data analysis and utilization
E. Statistical research and training aspects
1. Infrastructure
2. Courses offered (type and location)
3. Attendance and output
F. Existing infrastructure of the statistical agency
1. Organization and structure
2. Statistical legislation
3. Existence of statistical advisory/policy councils
4. Civil service conditions
G. External assistance
1. Multilateral assistance
2. Bilateral assistance
3. Non-governmental assistance
4. Source and extent of funding for statistical work
5. Coordination of assistance
H. Recommendations arising from the above evaluations
1. Future statistical work plan
2. Statistical priorities
3. Statistical legislation
4. Statistical advisory board (or similar)
5. Organizational structures
6. Other
Annexes on budgets, staffing, organization charts, lists of statistical output, etc.