Survey Statistician · Sampling Consultant · Data Scientist

Expert survey design and sampling methodology for organizations that need results they can trust

I assist international organizations, government agencies, and research institutions design rigorous household surveys, build in-country statistical capacity, and produce reliable subnational estimates — from inception to final report. I provide short-term technical assistance and multi-year programs, in English or French.

Available for engagements — remote & on-site worldwide
Survey Design Sampling & Estimation Small Area Estimation Capacity Building Causal Inference Data Science Statistical Software Python R Stata
Work with me → See selected work
Mamadou S. Diallo, Ph.D. — Survey Statistician and Sampling Consultant based in Hoboken, NJ
20
Years in complex survey design and analysis
195
Countries (WUENIC, UNICEF)
20+
Peer-reviewed publications
30+
Countries capacity-built
UNICEF Westat NORC Statistics Canada United Nations WHO Gavi NSOs

About

Statistician by training.
Builder by instinct.

I'm Mamadou S. Diallo — a Ph.D. statistician and independent consultant based in Hoboken, NJ. For nearly two decades, I have helped international organizations, government agencies, and research institutions design surveys that produce credible, decision-ready evidence. My clients include Statistics Canada, UNICEF, Westat, and NORC at the University of Chicago.

My applied work spans leading the global WUENIC immunization coverage estimates for 195 countries at WHO/UNICEF, designing multi-stage national household surveys across Sub-Saharan Africa, and training national statistics offices — from Ethiopia to Vietnam — in advanced sampling and small area estimation techniques.

I believe rigorous methodology should be accessible to every researcher. That's why I build open-source tools — Samplics, svy, svyLab — that put production-grade survey statistics within reach of teams everywhere, regardless of infrastructure.

Education

Ph.D. Statistics, Carleton University
M.Sc. Statistics, Laval University

Credentials

ASA Member · AAPOR Member · Guest Editor, JSSAM 2025

Languages

English & French — fluent in both speaking and writing

Based in

Hoboken, NJ — available for remote and on-site engagements worldwide

Currently accepting engagements — short-term technical assistance, survey design mandates, and capacity building programs. Bilingual English/French.

Services

How I can help

I work with international organizations, government agencies, NGOs, and research institutions on engagements ranging from short-term technical assistance to multi-year survey programs — in English or French, remotely or on-site worldwide.

Household Survey Design & Sampling

End-to-end design of complex, multi-stage household surveys — sample frame development, sample size calculation, stratification, cluster selection, weighting, calibration, and variance estimation. Compliant with international standards for DHS, MICS, LSMS, and custom survey programs.

Statistical Capacity Building

Hands-on training and long-term technical backstopping for national statistics offices, research teams, and survey practitioners — delivered in English or French, remotely or on-site. Topics include sampling design, weighting, small area estimation, and reproducible analysis workflows.

Data Science & Survey Analysis

Rigorous analysis of complex survey and health data — from weighted cross-tabulations and regression to machine learning and causal inference. Delivered with reproducible Python and R workflows that your team can maintain and build on.

Small Area Estimation (SAE)

Reliable subnational and disaggregated estimates when direct survey samples are too small — using model-based SAE methods (Fay-Herriot, unit-level, spatial) refined through years of applied work with national statistics offices across Africa and Asia.

Statistical Software & Tool Development

Custom Python and R tools for survey operations — CAPI-ready data pipelines, sampling selection utilities, automated weighting systems, analysis dashboards, and reproducible reporting. Built to be maintainable by your team after the engagement ends.

Impact Evaluation Design

Quasi-experimental and randomized designs for measuring program impact in education, nutrition, health, and livelihoods — across low- and middle-income country contexts. From power calculations and sampling strategy through data collection oversight and final analysis.

Work & Tools

Selected projects & open-source tools

A selection of the tools I build and the work that shapes them.

Platform

svyLab™

A cloud-native, AI-enabled analytics platform unifying survey, epidemiological, and health data workflows. Built on Google Cloud Run with CI/CD, svyLab brings professional-grade survey analysis to researchers who need reproducible, shareable results — without managing infrastructure.

Visit svylab.com →
Python Package

svy

A high-performance Python package for survey design and analysis and design-based causal inference. Built with Rust, Polars, and msgspec under the hood for speed and reliability.

View on PyPI →
Python Package

Samplics

A comprehensive open-source library for sampling, weighting, and analysis of complex survey data. Published in the Journal of Open-Source Software (JOSS).

JOSS Paper →
Consulting

McGovern Dole Program Surveys

Led the end-to-end design of national education and nutrition surveys in The Gambia (2023) and Togo (2024) for Catholic Relief Services — from sample design to data analysis.

Research

PHIA Surveys — Sub-Saharan Africa

Lead sampling statistician for the Population-based HIV Impact Assessment surveys across eight African nations at Westat — a flagship official statistics initiative.

Research

WUENIC — Global Immunization Estimates

Led the annual production of WHO/UNICEF Estimates of National Immunization Coverage for 14 vaccines across all 195 UNICEF member states — integrating survey, administrative, and geospatial data.

Capacity Building

Small Area Estimation for NSOs

Built technical capacity in SAE methods for the national statistical offices (NSOs) of Burundi, Ethiopia, Senegal, and online trainings for many more NSOs — enabling subnational estimates from existing survey data.

Publications

Selected publications

Peer-reviewed work in survey methodology, small area estimation, public health, and machine learning.

View all publications on Google Scholar →

Get in touch

Let's work together

I'm available for consulting engagements — survey design mandates, sampling methodology review, small area estimation, statistical capacity building workshops, and long-term technical assistance programs. I work in English and French, remotely or on-site worldwide.

Typical clients include UN agencies, bilateral donors, NGOs, national statistics offices, and academic research teams. I respond to all inquiries within two business days.

✉
Email
msdiallo@samplics.org
🌐
Platform
svylab.com
in
LinkedIn
linkedin.com/in/MamadouSDiallo
📄
Publications
Google Scholar

Send a message

I'll respond within two business days.