Aswesuma — Sri Lanka's Algorithmic Welfare-Targeting Score, and Its Exclusion Problem
Sri Lanka
Sri Lanka's Aswesuma programme uses a 22-indicator Multi-dimensional Deprivation Score computed by the Department of Census and Statistics to sort …
Colombia · Bogotá · See the Colombia profile · See the Bogotá profile
Top 75% 47/100 · Ask Evidence Copilot about this practice
Colombia's Sisbén IV algorithm scores households on housing, health, education and labour data to decide eligibility for social programmes nationwide. A 2024 study found only slight agreement with the official poverty index: 37.69% of non-poor households misclassified as poor.
Sisbén (Sistema de Identificación de Potenciales Beneficiarios de Programas Sociales) is Colombia's national targeting instrument, run by the Departamento Nacional de Planeación (DNP), used to decide who qualifies for dozens of social programmes: health subsidies, the Colombia Mayor old-age pension, housing subsidies, ICBF childcare and education grants among them. Since 2020, Metodología IV has combined housing, health, education and labour-market variables into a household score, replacing the earlier pure per-capita-income proxy and sorting households into four groups (A-D) with subgroups that individual programmes use to set their own eligibility cut-offs.
A peer-reviewed 2024 study in Desarrollo y Sociedad (Universidad de los Andes) directly compared Sisbén IV's classification of households against Colombia's official Multidimensional Poverty Index (IPM) and found only slight statistical agreement (Cohen's kappa below 0.36): 37.69% of households the IPM does not consider poor were nonetheless classified as poor by Sisbén, while 4.89% of IPM-poor households were missed entirely. Independent analysis (Guberney/Al Poniente, 2026) situates these figures against known problems in the prior methodology, Sisbén III, which earlier studies found had inclusion errors of 49.9% against monetary poverty and 64.8% against multidimensional poverty.
DNP's public operating manual documents the full points-based methodology, and the government presents Metodología IV as an improvement on its predecessor. But the same literature that documents the improvement also finds the current score still substantially misaligned with the government's own poverty measure, with real consequences: misclassified households can be wrongly denied or wrongly granted access to means-tested benefits. Advocacy and academic analysis have raised concerns about the opacity of the score to the citizens it classifies, and about the dignity implications of an algorithmic gate to essential services.
Read the full analysis: https://www.sdp.gov.co/gestion-estudios-estrategicos/sisben/metodologia-4
Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.
Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.
Where this practice's information was retrieved from, and when.
Sri Lanka
Sri Lanka's Aswesuma programme uses a 22-indicator Multi-dimensional Deprivation Score computed by the Department of Census and Statistics to sort …
Azerbaijan
Azerbaijan's State Social Protection Fund uses AI-based data matching across 15+ agencies to proactively identify and enrol eligible citizens in …
Kazakhstan
Kazakhstan's Digital Family Card uses predictive scoring across 100+ socio-economic indicators from 20+ agencies to proactively identify and serve vulnerable …
Chile
Chile's RSH algorithm classifies 8.99 million households (85.6 % of the population) into socioeconomic tiers that gate access to all …
Open full copilot Grounded in cited practices — always check the sources.