International Individual Consultant: To Develop Small Area Spatial Modelling for district level (admin-2) indicators in Mozambique

Terms of References

Hiring Office: UNFPA Mozambique Country Office

Purpose of consultancy: Develop Small Area Geospatial Modelling for district level (admin-2) indicators in Mozambique using Demographic & Health Survey 2022-23

Scope of work: (Description of services, activities, or outputs): 

A better understanding of geographic variation and inequities in demographic and health outcomes within countries is increasingly important for public health efforts.
Disaggregating data to the small area level is widely recognized as central to achieving the Sustainable Development Goals (SDGs). Many developing countries,
including Mozambique, have adopted a decentralized approach to governance for effective delivery of health care and other social services. In this context, small areas
estimates - usually at the lower subnational administrative level (Admin 2 or district) have become increasingly important for policy making, resource allocation,
program monitoring and evaluation. However, in some countries, small area estimates are available only for indicators derived directly from the population census, which provides limited information on socio-economic and demographic indicators. Key demographic and health indicators, such as contraceptive prevalence, nutrition and fertility are usually not covered by the census.

Nationally representative surveys, such as the Demographic and Health Surveys (DHS), produce reliable estimates of demographic and health indicators at the
national level and the first subnational administrative level (Admin 1: provinces, states, or regions). Because national-level estimates are useful for comparing countries and aggregating data across large regions of the world, their natural audience includes international policymakers and donors (Li et al. 2019). While Admin 1 analyses are useful for understanding the distribution of health and demographic phenomena, they do not provide comprehensive estimates at lower levels, such as the second subnational administrative level (Admin 2), where health programs are designed and implemented (Li et al. 2019, Mayala et al. 2019). Notably, Admin 2 areas are often referred to as districts in many countries, including Mozambique.

While DHS surveys are more regular and collect a significant amount of subnational (Admin 1) representative data, they cannot directly generate reliable estimates of
indicators at the small area level due to small sample sizes, which result in high sampling variability. Three approaches are currently available in the literature to estimate population-based survey indicators at small geographical units.

1. Extending the process of collecting nationally representative survey data by increasing the sample size, survey costs and survey time required to produce a representative sample at the desired administrative level.
2. Using data from routine health information systems from health facilities or communities.
3. Small Area Estimation (SAE) - statistical modelling and spatial interpolation techniques to predict indicator values for small geographic units using survey data and geospatial covariates.
 

The first approach is often not feasible in an increasingly resource-constrained environment. The quality of data for the second option is not always reliable.
Additionally, the data is not easily accessible and usually not nationally representative. The third approach – small area estimation – has gained traction in
recent years as it offers cost effective and statistically sound means to generate small area estimates. Machine learning and Bayesian geospatial modelling techniques
have proven effective in leveraging survey data and geospatial covariates to predict demographic and health indicators at finer geographic scales.

UNFPA works continuously to generate high quality disaggregated data and analysis to strengthen national capacity for policy formulation, planning and monitoring, as well as advocacy and engagement in policy dialogues to position population dynamics as a core development concern in the national development framework. In partnership with the Ministry of Economy and Finance, Government of Mozambique, UNFPA is currently providing technical assistance and capacity building to stakeholders to improve the use of data in decision-making at the national and sub-national levels. In addition, as part of CPD10, UNFPA is prioritizing the generation of district - level (Admin 2) estimates of selected SDG indicators using data from the nationally representative household surveys.
 

Recognizing the increasing demand for small area estimates, the DHS program has launched spatially modelled Admin2 estimates, publicly available publicly available through the Local Data Mapping Tool (https://spatialdata.dhsprogram.com/local- data-mapping-tool/). The estimates are a useful tool in evaluating geographic disparities within subnational and national boundaries. When Admin 2 estimates are produced for multiple surveys over time, program planners can see how indicators change for individual Admin 2s, making the estimates a useful tool for monitoring and evaluating subnational progress.

Duration and working schedule: The consultancy is for a period of 45 working days. The consultancy will start from (July 01, 2025)

Place where services are to be delivered: The consultant will work remotely on the given assignment. However, he will travel to Maputo to facilitate the workshop with national stakeholders.

Required expertise, qualifications and competencies, including language requirements: 

The Consultant should have extensive experience in Geospatial information system, data analytics, machine learning and geo-statistical modelling. Proven record in leading multidisciplinary teams and developing innovative predictive models, machine learning algorithms, and geospatial models to inform decision-making and address public health challenges. Skilled in managing and developing projects. S/He will possess excellent analytical, communication, and writing skills. Previous experience in conducting analysis from the DHS will be considered an asset. His/her primary responsibilities will be:

The expected outputs and deliverables are as follows:

5. Inception report outlining the consultants’ understanding of the ToRs, and proposed list of indicators. The report should translate the requirements of the ToR
into a practical and feasible approach to the dashboard, and work plan.
a) To produce the small areas estimates of selected indicators at the district level in Mozambique.
b) Publicly available the standard set of spatially modelled map surfaces at the DHS official forum.
c) Develop the data visualization tools (R-shiny and others) for navigating the results Create data visualization tools (R-shiny and other platforms) to explore
the results.

Delivery dates and how work will be delivered (e.g. electronic, hard copy etc.): All material will be submitted to INE and UNFPA in soft/electronic form.

📚 𝗗𝗶𝘀𝗰𝗼𝘃𝗲𝗿 𝗛𝗼𝘄 𝘁𝗼 𝗚𝗲𝘁 𝗮 𝗝𝗼𝗯 𝗶𝗻 𝘁𝗵𝗲 𝗨𝗡 𝗶𝗻 𝟮𝟬𝟮𝟯! 🌍🤝 𝗥𝗲𝗮𝗱 𝗼𝘂𝗿 𝗡𝗘𝗪 𝗥𝗲𝗰𝗿𝘂𝗶𝘁𝗺𝗲𝗻𝘁 𝗚𝘂𝗶𝗱𝗲 𝘁𝗼 𝘁𝗵𝗲 𝗨𝗡 𝟮𝟬𝟮𝟯 𝘄𝗶𝘁𝗵 𝘁𝗲𝘀𝘁 𝘀𝗮𝗺𝗽𝗹𝗲𝘀 𝗳𝗼𝗿 𝗨𝗡𝗛𝗖𝗥, 𝗪𝗙𝗣, 𝗨𝗡𝗜𝗖𝗘𝗙, 𝗨𝗡𝗗𝗦𝗦, 𝗨𝗡𝗙𝗣𝗔, 𝗜𝗢𝗠 𝗮𝗻𝗱 𝗼𝘁𝗵𝗲𝗿𝘀! 🌐

⚠️ 𝐂𝐡𝐚𝐧𝐠𝐞 𝐘𝐨𝐮𝐫 𝐋𝐢𝐟𝐞 𝐍𝐨𝐰: 𝐏𝐨𝐰𝐞𝐫𝐟𝐮𝐥 𝐓𝐞𝐜𝐡𝐧𝐢𝐪𝐮𝐞𝐬 𝐡𝐨𝐰 𝐭𝐨 𝐠𝐞𝐭 𝐚 𝐣𝐨𝐛 𝐢𝐧 𝐭𝐡𝐞 𝐔𝐧𝐢𝐭𝐞𝐝 𝐍𝐚𝐭𝐢𝐨𝐧𝐬 𝐍𝐎𝐖!

Monitoring and progress control, including reporting requirements, periodicity format and deadline: The consultant will be in close coordination with UNFPA, an INE Senior management and other technical staff to ensure quality and standards.

Deliverables and payment schedule: The key deliverables and payment schedule are as follows:

Milestone 1: A draft report on the small area estimates of selected indicators at the district (Admin2) level using data from the 2022-23 DHS survey and geospatial covariates. Payment (condition upon approval) 50%.

Milestone 2: Conduct the in-country workshop on the methodology and reproducibility of the results for INE staff. Payment (condition upon approval) 25%.

Milestone 3: Develop the data visualization tools (R-shiny and others) for navigating the results. Payment (condition upon approval) 25%.

Supervisory arrangements: The Consultant will work under the direct supervision Technical Specialist (P&D Team Leader), UNFPA.

Expected travel: The consultant needs to travel to Maputo to facilitate the one-week workshop with national stakeholders. UNFPA will pay for most economic and direct flights from the home place to Maputo and pay the DSA according to UNFPA rules and regulations.

Required expertise, qualifications and competencies, including language requirements: 

Master’s degree or higher in Geospatial Sciences, Data sciences, or a relevant combination of education for technical consultation. Minimum 7 years of professional experience in the field of geospatial Information system, data analytics, machine learning and complex modelling, preferably in humanitarian context, health, education and Census/DHS. Knowledge of the DHS indicators and socio-cultural issues. Experience in developing geospatial models to inform decision-making. Proven ability to work effectively with government officials. Languages: Fluency in oral and written English and knowledge of
Portuguese. Excellent interpersonal and general communication skills.  

Inputs / services to be provided by UNFPA or implementing partner (e.g support services, office space, equipment), if applicable: A consultant is expected to use his/her own laptop or any other equipment if required.

Other relevant information or special conditions, if any: Individual consultancy contracts will be signed between the Consultant and UNFPA Mozambique Country Office.  

UNFPA Work Environment:

UNFPA provides a work environment that reflects the values of gender equality, diversity, integrity and healthy work-life balance. We are committed to ensuring gender parity in the organization and therefore encourage women to apply. Individuals from the LGBTQIA+ community, minority ethnic groups, indigenous populations, persons with disabilities, and other underrepresented groups are highly encouraged to apply. UNFPA promotes equal opportunities in terms of appointment, training, compensation and selection for all regardless of personal characteristics and dimensions of diversity. Diversity, Equity and Inclusion is at the heart of UNFPA's workforce - click here to learn more.

Disclaimer:

Selection and appointment may be subject to background and reference checks, medical clearance, visa issuance and other administrative requirements. 

UNFPA does not charge any application, processing, training, interviewing, testing or other fee in connection with the application or recruitment process and does not concern itself with information on applicants' bank accounts. 

Applicants for positions in the international Professional and higher categories, who hold permanent resident status in a country other than their country of nationality, may be required to renounce such status upon their appointment.