LOADING
Needs Assessment Clustering and Predictive Modelling Pilot
Location: Remote
Estimated Level of Effort: 42 days
The International Rescue Committee (IRC) responds to the world's worst humanitarian crises, helping to restore health, safety, education, economic wellbeing and power to people devastated by conflict and disaster. Operating in over 40 countries and 20+ US cities, the IRC works at the intersection of humanitarian response and longer-term development, serving millions of clients annually. The IRC's METAL Unit leads the organisation's approach to monitoring, evaluation, technology, accountability and learning, supporting country programmes to collect, manage and use data to improve programming. The Unit maintains a suite of digital data collection and visualisation tools — including CommCare, Power BI and Databricks — and is increasingly investing in the use of advanced analytics to extract actionable insights from the large volumes of programme data the IRC collects.
The IRC recently completed a machine learning exercise using client registration data from its Ecuador-Peru country programme. Using K-medoids clustering on a dataset of approximately 13,000 clients, the analysis revealed four distinct population segments — including a critical finding that 41% of children and families in transit were receiving no services at all. This insight, invisible to conventional demographic analysis, has since shaped programme adaptations in both countries and informed donor proposals. The exercise validated a straightforward principle: machine learning can identify patterns in existing programme data that human analysis misses, enabling smarter targeting, stronger proposals, and more client-centred programming. The IRC now wishes to build on this proof of concept in two directions: applying clustering to a new needs assessment context, and developing a first predictive model for a specific programme intervention. This consultancy is scoped to deliver both workstreams as discrete Phase 1 outputs of the IRC's broader ML for Programmes initiative.
The consultant will:
Review available datasets — likely including client registration data, Multi-Sector Needs Assessment (MSNA) outputs, and programme monitoring data — and assess quality, completeness and suitability for clustering. Produce a brief data assessment note (2–3 pages) documenting sources used, limitations, and any assumptions made in preparation. Conduct necessary data cleaning and structuring. This note will serve as a formal sign-off point before analytical work proceeds.
Apply an appropriate clustering algorithm (eg K-medoids, hierarchical clustering, or similar) to identify distinct population segments. The choice of method should be justified and documented. In parallel, conduct a methodological assessment of whether MRP techniques can extend the reach of the analysis to data-sparse locations. The consultant should attempt an MRP model where data conditions permit, report the resulting estimates with explicit uncertainty intervals, and document clearly what the estimates show, the limitations of the approach given available auxiliary data, and what additional data collection would be required to improve reliability. The MRP component is understood to be exploratory; its value lies as much in surfacing the boundaries of current evidence as in producing point estimates.
Produce a written needs assessment report interpreting the cluster outputs in programmatic terms. The report should translate statistical findings into actionable insights — describing who the population segments are, what their distinct needs appear to be, and what the implications are for programme design and targeting. Where MRP estimates are included, they should be clearly labelled with appropriate caveats. The IRC's MEAL and programme staff will contribute to contextual interpretation and the final write-up; the consultant leads the analytical narrative.
Review candidate programme datasets — likely spanning two to three programme areas (potentially including economic recovery and development, education, or protection) — and assess their suitability for predictive modelling. Suitability criteria include: availability of historical outcome data, sample size, data completeness, and ethical considerations around the use of predictions in that programmatic context. Produce a brief feasibility note recommending which programme to proceed with and why, agreed with the IRC before model development begins.
Following agreement with the IRC on the selected programme, build a predictive model using historical programme data. The model should predict a clearly defined outcome (eg programme completion, dropout risk, or a specific client-level result) using data available at or shortly after client intake. Model development must follow recognised best practices throughout, including:
Translate the validated model into a lightweight operational tool — a scoring widget, dashboard integration, or structured output — that programme staff can use without specialist data science knowledge. Produce accompanying documentation covering: how the tool works, how to interpret its outputs, its known limitations, and recommended protocols for human oversight and override. The tool should be deployable within the IRC's existing Databricks and Power BI infrastructure.
Drawing on the experience and learning from both workstreams, produce a practical methodological guide (approximately 8–10 pages) intended for IRC Regional Measurement Advisers. The guide should be written for an audience that is quantitatively literate but not specialist data scientists. It should cover: how to assess whether available data is suitable for clustering or predictive modelling; rules of thumb for method selection; key questions to ask when commissioning or reviewing this type of work; how to sense-check analytical outputs; and how to present findings to programme and business development audiences. The guide should be grounded in the specific experience of this consultancy rather than generic methodology, and should be reviewed by at least one RMA before finalisation.
| Deliverable | Workstream | Timeframe | Estimated Days |
|---|---|---|---|
| Deliverable 1: Data assessment note | A | Weeks 1–2 | 3 |
| Deliverable 2: Data cleaning and preparation | A | Weeks 2–5 | 7 |
| Deliverable 3: Clustering analysis and MRP assessment | A | Weeks 5–9 | 8 |
| Deliverable 4: Needs assessment report, final | A | Weeks 9–12 | 5 |
| Deliverable 5: Data feasibility assessment note | B | Weeks 2–4 | 3 |
| Deliverable 6: Model development and validation | B | Weeks 8–14 | 8 |
| Deliverable 7: Operational tool and documentation | B | Weeks 14–17 | 4 |
| Deliverable 8: RMA methodology guide | C | Weeks 15–18 | 4 |
| Total Estimated Days | 42 | ||
Note: Workstreams A and B run in parallel. The timeline assumes data access is provided promptly at inception and that the data assessment sign-off (Workstream A) and feasibility note agreement (Workstream B) proceed without significant delay. Delays in data access or sign-off will affect the schedule accordingly.
The IRC takes data protection seriously and operates under its organisation-wide data protection policy. The consultant will be required to sign a data sharing agreement prior to accessing any client-level data. All data provided will be de-identified. The consultant must not share, publish or retain IRC data beyond the terms of this agreement. Predictive models built under this ToR are intended solely to support the allocation of additional resources and tailored support — not to exclude clients from services or make automated decisions affecting individual cases. The robustness checks and subgroup performance analysis described in section 4B are a mandatory component of the work, not optional enhancements. The consultant is expected to engage with these ethical parameters from the feasibility assessment stage onward, and to reflect them explicitly in the tool documentation and any outputs shared with IRC programme staff.
The consultant will report to [Philip Blue / Regional Measurement Adviser, Latin America] and work in close coordination with the IRC's Measurement Unit and relevant country programme MEAL staff. An inception call will be held within the first week to align on data access, workplan, and communication rhythms. Brief fortnightly check-ins are expected throughout. Two formal review points are built into the timeline: sign-off on the Workstream A data assessment note, and agreement on the Workstream B feasibility recommendation, both before substantive analytical work proceeds. The Workstream C methodology guide will be reviewed by at least one Regional Measurement Adviser in addition to the primary supervisor prior to finalisation.
All International Rescue Committee workers must adhere to the core values and principles outlined in IRC Way - Standards for Professional Conduct. Our Standards are Integrity, Service, Equality and Accountability. In accordance with these values, the IRC operates and enforces policies on Safeguarding, Conflicts of Interest, Fiscal Integrity, and Reporting Wrongdoing and Protection from Retaliation. IRC is committed to take all necessary preventive measures and create an environment where people feel safe, and to take all necessary actions and corrective measures when harm occurs. IRC builds teams of professionals who promote critical reflection, power sharing, debate, and objectivity to deliver the best possible services to our clients.
Cookies: https://careers.rescue.org/us/en/cookiesettings
IRC is an Equal Opportunity Employer. IRC considers all applicants on the basis of merit without regard to race, sex, color, national origin, religion, sexual orientation, age, marital status, veteran status, disability or any other characteristic protected by applicable law. The International Rescue Committee (IRC) helps people whose lives have been shattered by conflict and disaster to survive, recover and rebuild. We are trusted for our ability to reach people in the toughest places, put our clients at the center of everything we do, and - as an evidence-based organization - to relentlessly improve the impact we deliver for those we serve. If you're ready to be part of a team that brings ingenuity, fortitude and optimism to making a lasting impact, we want to hear from you. The IRC Way, our code of conduct, is founded upon our core values of Integrity, Accountability, Service and Equality. These values unite IRC staff around the world and keep our programs firmly directed towards the well-being of our clients while meeting the expectations of our donors as responsible stewards of their resources.