Hardship Level (not applicable for home-based)A (least hardship)

Family Type (not applicable for home-based)

Family

Staff Member / Affiliate TypeUNOPS IICA1

Target Start Date2024-08-01

Job Posting End DateJuly 30, 2024

Terms of Reference1. General Background

UNHCR's Data Transformation Strategy 2020-2025 aims to establish the organization as a trusted leader on refugee and displacement data by enhancing data management, innovative systems, skill development, and evidence-based decision-making.

The Data Science and Curation Specialist will leverage cutting-edge data science methods like AI, machine learning, remote sensing, and predictive analytics alongside advanced data curation practices. Responsibilities include improving data analysis and dissemination, enhancing data systems, managing data retention, and ensuring data availability through UNHCR’s data libraries. The role supports both global and regional decision-making by making anonymized data accessible for analysis, improving data quality and institutional memory.

Reporting to senior DIMA coordinator and the Statistics and Data Analysis Officer within the Regional Bureau, this position involves collaboration with internal and external stakeholders. The Data Science and Curation specialist plays a key role in implementing UNHCR's vision for data-driven empowerment and protection of displaced individuals.

2. Purpose and Scope of Assignment

The Data Science and Curation Specialist will be responsible for:

assisting in develop predictive model for forced displacement,

exploring the use of emerging technologies for the analysis of forced displacement data,

assisting with data selection and assessing the suitability of the data for publication,

cleaning data

identifying and addressing missing data,

applying imputation methods as needed,

anonymizing personally identifiable data,

populating the microdata library with datasets and corresponding metadata,

training of staff in countries to decentralized data curation activities,

coordinating of data curation in countries.

The Data Science and Curation Specialist will more specifically perform the following activities:

Apply data science techniques to regional forced displacement and statelessness situations to improve understanding of population behavior, needs, and vulnerabilities and capacity to monitor and evaluate the impact of UNHCR policies.

Liaise with prioritized country operations within the region to identify and obtain historical datasets suitable for sharing.

Work with country offices to have data stored within RIDL (Raw Internal Data Library) according to data security and data protection protocols.

Work with field offices and partners to ensure relevant metadata is provided along with the data.

Develop instructions and guidance documents to evaluate the quality of the selected datasets and assess suitability for publication, removing and correcting errors if necessary, and work with field locations as relevant.

Assist in the development of communications materials and activities to promote the sharing and use of micro datasets.

Apply statistical disclosure control (SDC) techniques used to anonymize all datasets ensuring that no person or household is identifiable from the results of an analysis of survey data (at an appropriate level of risk), or in the release of microdata, utilizing the R-based sdcMicro package. This will be done in consultation with the data producer that will guide in the determination of key and sensitive variables. Such anonymization will be performed both at the household level and individual level and applying different methods for public use and licensed use files. Perform statistical analysis on pre vs post anonymization utility variables to verify anonymized data maintains validity and should be recommended for sharing.

Generate the appropriate metadata according to the Data Documentation Initiative (DDI) metadata standard and any other documentation needed to catalogue, record, and communicate to users all the salient characteristics of the micro-datasets shared on the MicroData Library.

Undertake missions to country offices and select field operations in his/her reference region to discover data and train and assist country staff in the use of the platforms.

Assist in the organization and/or delivery of curation workshops at the regional and country level.

Coordinates data curation activities at country level in operations

Contribute to the research on the use of alternative data sources to produce data on regional forced displacement and statelessness situations.

Facilitate open access to anonymized forced displacement data while addressing protection and privacy concerns over microdata managed by UNHCR.

Advanced data quality assurance approaches in UNHCR, bringing them in line with international standards in data quality.

Apply predictive analytics techniques to produce population statistics.

Perform other related duties as required.

3. Monitoring and Progress Controls

Develop and publish/put in product predictive models, at least one model.

RIDL to be populated at a minimum with all raw data used to populate the Microdata library.

The Microdata library to be populated with at least 15 clean, anonymized, and appropriately documented datasets by the end of the assignment. Each data set will ideally be from a different survey. Should that not be feasible, more than one dataset from the same survey may be acceptable if it refers to a separate country or a different refugee population and a different period.

Additional datasets related to protection monitoring systems, programme monitoring systems, needs assessments, shape files for mapping and other data systems that relate to the population of concern. will be indexed in RIDL.

Drawing from the experience gained in populating the data library, establishment of guidance materials regarding cleaning and anonymization of selected datasets.

Support with the institutionalization of curation activities in the regional DIMA and with training of relevant staff in field operations.

Generation of scripts and automated processes for the extraction and processing of standardized data and the creation of anonymization reports.



4. Qualifications and Experience

a. Education (Level and area of required and/or preferred education)

• University Degree (Master’s or PhD equivalent) in Statistics, Computer Science, Applied Mathematics, Economics, Demography, Quantitative Social Sciences, or related field.

Standard Job Description

Required Languages

,

,

Desired Languages

,

,

Additional Qualifications

Skills

Education

Certifications

Work Experience

Other informationThis position doesn't require a functional clearance


Home-BasedNo

Recommended for you