Resilience Index Measurement and Analysis (RIMA) Survey in Afghanistan, Second Round, 2024

A two-stage cluster sampling approach has been applied to select households for interview. In the first stage, an equal number of clusters per province has been selected using Probability Proportional to Size (PPS). In the second stage, a fixed number of households per cluster has been randomly selected among households that participated in one or more of the following interventions: wheat cultivation packages, livestock protection, cash-for-work, unconditional cash transfers, poultry, small farm equipment, summer crops, and home gardening.

Household resilience has been assessed by statistically testing for differences in the means of key outcome variables between beneficiary households and non-beneficiary (control) households. Control households have been randomly selected from households that did not receive any assistance-neither from FAO nor from other organizations-during the project period. The sampling frame for control households included only households that met the targeting criteria for FAO interventions, in order to ensure that beneficiary and control households (i) had similar socioeconomic characteristics and (ii) were located in the same geographical areas.

For beneficiary households, sampling was based on complete beneficiary lists from the projects under analysis, from which a random sample of households has been drawn. For control households, a list of comparably vulnerable households was compiled in the selected provinces based on the eligibility criteria of the three projects, and a random sample has been drawn from these lists. The sampling process was documented.

Households were surveyed across all 34 provinces. An oversampling rate of 30 percent was applied to account for attrition across successive survey rounds aimed at tracking the same households over time.