Result of Service

By 30 September 2023
• Concept note, and implementation schedule for the delivery of technical support and training, indicating the timeline of activities from the start of the assignment and after the delivery of the training for implementing the methods and providing feedback on the use of SAE methods.
• Report, in English, on progress on development of training materials, and remote technical support provided to identify and prepare datasets, including details on:
o Different survey, census, administrative and geospatial datasets, including relevant auxiliary variables, that will be used as potential inputs, in order to identify appropriate methods to focus on in the training
o Profile of the participants defined in collaboration with the GSO
o Tools, software, data and analytical environment needed to deliver the training, including the characteristics of any synthetic datasets, if appropriate, in consultation with the GSO
o Outline for the training agreed in consultation with the GSO and UNSD

By 30 October 2023
• Prepare and deliver a comprehensive, training course (Class 1), to be delivered through a 5-day in person delivery to national staff, covering:
o area-level small area estimation models, including Fay-Herriot and relevant variants;
o unit-level small area estimation models such as ELL, Molina-Rao, including extensions to cover the use of geospatial data (e.g. Masaki et al., 2020);
• A summary report of the outcome of the training

By 30 December 2023
• Prepare and deliver a in-person training sessions (Class 2) with staff from GSO and relevant line ministries. The training will be at least 5 days, and the schedule will be determined in consultation with the GSO, covering:
o Refresher on relevant area-level and unit-level SAE techniques as necessary;
o Practical application of relevant SAE methods to support estimation of selected MDCP and food insecurity indicators using GSO datasets (e.g., household surveys and censuses, georeferenced datasets) and publically available geospatial and population datasets, to provide relevant auxiliary variables covering
o Spatial techniques (e.g. INLA) for producing geographically disaggregated estimates.
o Estimation of uncertainty around modelled estimates and comparison with direct estimates;
o Benchmarking to align small area estimates to published direct estimates for higher levels of geography;
o Assessment of quality and evaluation of modelled estimates and selection of preferred model(s); and
o Appropriate communication and dissemination of modelled estimates.
• A summary report of the outcome of the training

By 29 February 2024
• Provide copies of presentations, exercises and any other materials, in English and Vietnamese, for GSO and for upload to the Data For Now website.
• A final report, in English and Vietnamese on the follow-up technical support provided and guidance to the GSO on revised versions of models and outputs produced for MDCP and food insecurity by the GSO technical team, a summary of the outcomes of the trainings, including details of the models produced, recommendations on the next steps for the implementation of SAE methods in Viet Nam, and short reflection piece on the training/SAE uptake in GSO.

Work Location

The consultant will be based in Hanoi, Viet Nam. The consultancy is home-based with regular phone and web conference meetings with staff members of GSO, UNSD and other Data4Now stakeholders, as well as in-country visits to partners and government offices. The consultant must be available for regular technical meetings to monitor the progress of the work with the GSO and the UNSD team. Travel or commute time to and from United Nations Headquarters, as well as related expenses, are not part of the consultancy.

Expected duration

The consultant will work in the period between 1 September 2023 and 29 February 2024. The payment is calculated in accordance with the national UN salary scales.

Duties and Responsibilities

The Data For Now initiative (Data4Now), co-led by the United Nations Statistics Division (UNSD), the World Bank, the Global Partnership for Sustainable Development Data (GPSDD), and the Sustainable Development Solutions Network (SDSN), aims to develop countries’ capacities to deliver the information needed by local and national policy and decision-makers to achieve the 2030 Agenda. To this end, it supports members of the national statistical systems in participating countries to collaborate more effectively with local, national, and global partners from intergovernmental organizations, academia, civil society, and the private sector.

As part of the implementation of the Data4Now initiative, the project “Accelerating implementation of Data4Now in eight countries in Africa and Asia” funded by the Italian Government, will support, through various trainings and advisory consultancies, the development of capacities to (1) use innovative sources, technologies and methods for the streamlined production and dissemination of better, more timely and disaggregated data for sustainable development and (2) make analytic insights derived from them readily available to policy and decision-makers.

The General Statistics Office (GSO) of Viet Nam has identified the production of estimates of child multidimensional poverty (MDCP) and food insecurity indicators disaggregated at the provincial or district level as two key priorities. In this context, Data4Now is seeking the support of a consultant to provide technical support and training to the relevant national staff in the application of small area estimation (SAE) for creating geographically disaggregated estimates of their priority indicators.

Under the guidance of teams from GSO and UNSD, the consultant is expected to provide technical support in identifying and preparing datasets for the application of small area estimation in Viet Nam, explore alternate data sources for integration as ancillary data (if available), provide training to strengthen capacity of relevant national staff to implement and adapt existing methods to produce small area estimates, and feedback on the use of SAE methods.

Specifically, the work assignment will involve the following:

1. Liaise with colleagues in the GSO to identify training needs, and assess current knowledge and experience of relevant estimation techniques, and review existing materials to be used in developing the training and develop bespoke materials on the implementation of small area estimation methods.

2. Develop the concept note, and implementation schedule for the delivery of technical support and training for the GSO and relevant government members, indicating the timeline of activities from the start of the assignment and after the delivery of the training for implementing the methods and providing feedback on the use of SAE methods, in consultation with GSO. Two trainings classes are expected, Class 1 is basic for a broader audience, and Class 2 is advanced for a selected number of trainees who are focused on producing estimates of the two indicators, MDCP Rate and Food Insecurity Rate, at provincial or district level for GSO to review and publish.

3. Provide support to GSO in identifying and preparing datasets, including identification of relevant auxiliary variables (both from GSO and open source earth observation/remote sensing data), for the application of small area estimation methods, and ensure actual and/or synthetic datasets are available in a suitable analytical environment for conducting the training and supporting GSO’s subsequent development of estimates.

4. Develop and deliver training, in Vietnamese, building on the technical knowledge already installed in GSO and taking into account the characteristics of the proposed data sources in identifying relevant methods. Focus should be on supporting GSO in applying these methods to create initial estimates at the province level for the food security and child multidimensional poverty variables of interest. Subject to the identified needs, these trainings should support the establishment of solid knowledge and practical skills from an applied standpoint in areas, using relevant software, including:
a) Use of area-level small area estimation models, including Fay-Herriot and relevant variants;
b) Use of unit-level small area estimation models such as ELL, Molina-Rao, Empirical Best/ Bayes including extensions to cover the use of geospatial data (e.g. Masaki et al., 2020);
c) Use of spatial techniques (e.g. INLA) for producing geographically disaggregated estimates;
d) Estimation of uncertainty around modelled estimates and comparison with direct estimates;
e) Use of benchmarking to align small area estimates to published direct estimates for higher levels of geography;
f) Assessment of quality and evaluation of modelled estimates and selection of preferred model(s); and
g) Appropriate communication and dissemination of modelled estimates.

5. Provide presentations and all support trainings materials in Vietnamese and English to UNSD for the GSO and upload to the Data For Now website.

6. Provide follow-up support to the training participants in their implementation of the techniques to produce provincial or district level department level estimates of MDCP and Food Insecurity. This should include responding to any queries raised by the GSO technical team and working to identify solutions for any practical issues identified in the application of the methods.

7. Provide feedback and guidance on revised versions of models and outputs produced by the GSO technical team, and recommendations on next steps for the implementation on the Small Area Estimation methods in GSO.

8. Share experiences on the use of SAE with other countries involved in the Data For Now project.

Qualifications/special skills

Advanced university degree (Master’s degree or equivalent) in economics, statistics, mathematics, environmental sciences, social sciences or a related field. A Bachelor’s degree in combination with two additional years of experience may be accepted in lieu of an advanced degree.

A minimum of ten (10) years of experience in statistical/analytical methods, data analysis and/or data management is required. Experience working with Stata, R, and/or Python is required. Experience working with National Statistical System is desirable. Experience working with innovation methods, tools and data is desirable. Experience working with data disaggregated by geography, sex and specific population groups of interest for sustainable development is desirable.

Languages

Full command of spoken and written Vietnamese and English is required.

No Fee

THE UNITED NATIONS DOES NOT CHARGE A FEE AT ANY STAGE OF THE RECRUITMENT PROCESS (APPLICATION, INTERVIEW MEETING, PROCESSING, OR TRAINING). THE UNITED NATIONS DOES NOT CONCERN ITSELF WITH INFORMATION ON APPLICANTS’ BANK ACCOUNTS.