Table of contents
Introduction 3
Opening addresses 3
Objectives of the Workshop 4
MDG monitoring at the sub-national level and administrative data 5
Discrepancies between national and international data 6
Net enrolment ratio and literacy 6
Poverty 6
Water and sanitation 7
Child mortality 8
Working groups and plenary discussion 8
NSS coordination and data reporting 10
The New MDG Monitoring Framework 12
Other issues 13
Recommendations 13
Annex 1. List of participants 16
Annex 2. Session on sub-national monitoring and administrative data: Summary of the Working Groups 22
Annex 3. Session on discrepancies between national and international data: Summary of the Working Groups 25
Annex 4. Summary of gaps and discrepancies between national and international data 29
4
Workshop on MDG Monitoring, Kampala, Uganda, 5-8 May 2008 / 2008Introduction
1. The Workshop on the Millennium Development Goals (MDG) Monitoring was held in Kampala on 5-8 May 2008. The workshop was organised by the United Nations Statistics Division (UNSD) in collaboration with the United Nations Economic Commission for Africa (UNECA). It was hosted by the Ugandan Bureau of Statistics (UBOS).
2. Participants included representatives from 17 National Statistics Offices - namely Botswana, Cameroon, Ghana, Kenya, Lesotho, Liberia, Malawi, Mauritius, Mozambique, Namibia, Norway, Sierra Leone, Swaziland, Tanzania, Uganda, Zambia and Zimbabwe -, from international agencies - United Nations (UN), United Nations Educational Scientific and Cultural Organisation (UNESCO), United Nations Children’s Fund (UNICEF), World Bank - as well as local participants from Department for International Development (DFID), Ministry of Local Government and Makerere University. The list of participants is available in Annex 1.
3. The Deputy Director of Statistics, Mr. Male Mukasa, welcomed the participants to Uganda and to UBOS. Highlighting on the importance of NSOs in monitoring development goals in general, he further stated that monitoring was central in achieving the MDGs. He noted that only eight years remained to the 2015 MDGs target and that everyone needed to work hard in full throttle in order to achieve the monitoring objectives. He concluded by wishing everyone a productive workshop.
Opening addresses
4. Three opening addresses were given. The first address was given by Mr. Ben Kiregyera, Director for African Centre for Statistics, who noted that much of Africa was in danger of failing to achieve all the MDG. He commended on the strides that had been taken to make statistics one of the clusters imbedded within the thematic clusters for scaling up interventions as identified by the MDG Africa Working Group. Noting the rise in statistics’ profile globally, he stated that statistics were being used to drive the outcomes that policies were meant to achieve. This in itself was a significant move from the traditional role that consisted only of the measurement of outcomes. In contrast to the global picture, he noted the immense challenges that were still facing African policy development where statistics were still not being mainstreamed into planning, policy and budget development. Prior to concluding, he noted the discrepancies between national and international data and the importance of resolving those differences. In conclusion, he saluted UNSD in providing technical leadership that had resulted in improved products and services that were currently being offered by the National Statistics offices (NSO).
5. The second address was given by his Excellency Mr. Percy W. Misika, Food and Agriculture Organisation Representative of the United Nations, on behalf of the UN Resident Coordinator. He began by noting how the adoption of the Millennium Declaration by the UN General Assembly had marked the dawn of a new millennium in ensuring global cooperation. He said that the review at the World Summit in 2005 had indicated that countries needed to triple their efforts in order to achieve these goals. He further went on to say that MDGs are technically feasible but that the current situation in some countries constrained the effectiveness and achievement of these goals. He stated that tracking MDG goals would assist policy makers, civil society and governments in developing appropriate socio-economic policies. He then proceeded on to state that achieving the MDG goals in Africa held the promise of saving many lives and added that the availing of good statistics would be central in ensuring success. He concluded by calling on countries to go beyond mere interest in regards to the application of statistics tools to meeting the monitoring tools.
6. The third and final address was given by Mr. Ackim Jere, representative of the SADC Secretariat. In unison with the previous two speakers, he welcomed all participants to the workshop. He then gave a brief overview of the SADC Development Account where UNSD is the executing agency. Prior to thanking UBOS for providing the venue, he called all SADC countries to attend an addendum after the workshop to review the progress made on the Development Account.
7. Ms. Francesca Perucci, on behalf of UNSD, welcomed everyone to the workshop noting the huge progress that had been accomplished as well as the progress needed to be made in the future. She stated that MDGs were now well established despite changes in the monitoring framework, systematic tracking was essential in sustaining development and reiterated the African Statistical Commission’s concern regarding the capabilities of the countries in producing the data that was needed. She stated the priority and commitment held by UNSD towards African countries. Ms. Perucci expounded on the key role of coordination in order to improve credibility and quality of the statistics produced. She alluded to the work done by the IAEG in addressing the differences between international and national data and stated that countries needed to work together to improve data quality and credibility.
Objectives of the Workshop
8. This workshop is one of the first initiatives to implement recommendations made by the international agencies and the countries to improve the monitoring of MDG indicators. The objectives of the Workshop are:
· To develop guidelines on how to improve coordination of reporting mechanisms;
· To review existing discrepancies between national and international data, identify data gaps at the international level and develop recommendations to address the gaps and discrepancies;
· Review methodological issues in MDG monitoring at the national and sub-national levels;
· Present the new targets and indicators that have been recently adopted, and review the corresponding metadata.
MDG monitoring at the sub-national level and administrative data
9. The session on MDG monitoring at the sub-national level started with four presentations:
MDG data at the sub-national level: relevance, challenges and IAEG recommendations (by UNSD).
This presentation informed the Workshop about the recommendations on MDG sub-national monitoring of the IAEG on MDG Indicators, outlined the relevance and the challenges of sub-national monitoring for countries, discussed the pros and cons of each data source, reviewed common approaches to combine data sources and briefly listed some advantages of using GIS for sub-national monitoring.
Use of administrative data sources and production of sub-national data for MDG monitoring in African countries: challenges and opportunities (by UNECA-ACS).
This presentation introduced the MDG Mapper, an online tool developed by UNECA for comparing, with maps, progress among African countries towards achieving the MDGs. The MDG Mapper displays sub-national data, when available. The presenter illustrated the use of MDG Mapper for sub-national monitoring with an example of net primary enrolment data for provinces in Ethiopia and poverty rates in different areas within Ghana.
Use of census and administrative data in the monitoring and education indicators: an international perspective (by UIS-UNESCO).
The representative of UIS-UNESCO explained that the enrolment data came mostly from administrative data provided by countries from their annual school census conducted by the Ministry of Education (UIS has data for 140 countries) while literacy data tended to be collected by NSOs in population censuses.
Mauritius: The MDGs-country experience (by Central Statistics Office of Mauritius).
The presenter explained that most MDG indicators were not relevant at the sub-national level because Mauritius is a small country. Only two MDG indicators are monitored at sub-national level: indicators 7.8 and 7.9 (respectively, Proportion of population using an improved drinking water source and Proportion of population using an improved sanitation facility) are compiled for urban and rural areas. The data are obtained from housing censuses, with full coverage ensuring the reliability of the monitoring at the sub-national level.
10. After the presentations, the participants were divided into two groups to share experiences and put forward recommendations for improving MDG monitoring at the sub-national level. The conclusions presented by the two groups are shown in Annex 2. During the group work and the plenary discussions, the following key issues were noted:
1) National figures tend to mask sub-national areas where interventions are mostly needed.
2) Geological Information Systems (GIS) are a useful tool in handling geographically referenced data.
3) The collection of administrative data still presents a challenge for many African countries. The representative of Botswana explained how policies in Botswana had enhanced the collection of births data by restricting enrolment into schools and access to medical care to children with registered births. The delegate noted that whilst policy had enhanced the collection of data on births, Botswana still faced the challenge of collecting data on deaths.
Discrepancies between national and international data
11. At the session on discrepancies between national and international data, UNSD presented the recommendations by the IAEG and EGM on MDG indicators on resolving discrepancies between international and national data sources. It was clarified that resolving did not mean to eliminate discrepancies but rather to understand the reasons why these discrepancies arise. In some cases, this understanding may lead to the reduction of the data differences, while in others there may be good reasons why the national and international data are distinct. In this case, it is important to clarify the user about the reasons explaining the differences so that the user can make an informed decision on which data are appropriate to use.
12. Presentations were given by UIS-UNESCO, UNICEF, UNICEF-JMP and the World Bank explaining the process used by these agencies to compile data from national sources and to produce international estimates. The representatives of Kenya and Malawi then presented respectively for indicators on water and sanitation (MDG indicators 7.8 and 7.9) and for selected education indicators (MDG indicators 2.1 and 2.3) their findings regarding data discrepancies between national and international data.
Net enrolment ratio and literacy
13. In the UIS-UNESCO presentation, Mr. Said Ould Voffal (UIS-UNESCO) justified the need to produce estimates for enrolment data: since the data received from countries is not comparable, estimation is needed to obtain a set of comparable data. On the other hand, for literacy data discrepancies between national and international data are rare because UIS tends to use directly the data received from the country – discrepancies can only arise if different sources are used or if country data are missing at UIS (in which case, UIS sometimes produces estimates using the GALP model). Mr. Said Ould Voffal also informed the participants that UIS consults with countries after cleaning the data and producing country estimates, but often countries do not reply.
14. Mr. Shelton Kanyanda (Malawi) listed the different data sources used in his country for the indicators 2.1 and 2.3. There are data discrepancies for both indicators. In Liberia, primary education consists of eight years. The international definitions define a shorter period of study for primary education and therefore the country data on indicator 2.1 (total net enrolment in primary) has been adjusted by UNESCO accordingly to maintain international comparability. For indicator 2.3, UNESCO is using different sources: the Malawi NSO provided data from the 1998 and 2005 Integrated Household Surveys while the international MDG database shows data from the 1998 Population and Housing Census.
Poverty
15. Mr. Johan Mistiaen (World Bank) started his presentation by clarifying that it is pointless to discuss discrepancies for indicator 1.1 (Proportion of population below $1 (PPP) per day) because this indicator is not adequate for country monitoring. For this indicator, the discussion on discrepancies should focus on the differences in the data used by the World Bank to calculate the indicator and the data available at the national level. The World Bank has an economist assigned to every country who is responsible for gathering the country data and submitting them to the World Bank Headquarters. Mr. Johan Mistiaen indicated that the World Bank has received data from all countries represented at this Workshop, except for Mauritius and Liberia. The representative of Mauritius clarified however that data are available at the country level.
Water and sanitation
16. Mr. Rolf Luyendijk (UNICEF-JMP) explained in his presentation that UNICEF-JMP applies linear regressions to fit all country data points available in order to estimate the water and sanitation indicators (7.8 and 7.9) – except if only two points too close in time are available then the average of the points is used. He also indicated that JMP abandoned the use of administrative data reported by the national governments because the data were unreliable, showing high variability from year to year. Survey data tends to be more consistent and therefore from 1997 onwards, only census and survey data have been used at the international level. Whenever the data values are not consistent, UNICEF-JMP contacts the country authorities to investigate the reliability of the data.
17. In her presentation on indicators 7.8 and 7.9, Ms. Mary M. Wanyoni (Kenya) showed that there are differences between the national and international data, especially for the sanitation indicator (40 versus 80). She explained the differences in definitions and methodologies used:
i. At the national level, improved drinking water sources do not include rain water (because it is rare to find households relying on rainwater) while this source is included at the international level; for sanitation, the international definition only considers the facility improved if it is not shared while the country does not use this criterion. The representative of Kenya indicated that the information on whether the sanitation facility is shared or not only started to be collected from 2003 onwards and requested the international agencies to distribute the methodology used to estimate access to non shared sanitation facilities for years prior to 2003.