Extended Data Table 2.
Analysis is similar to that presented in Table 1, but targeting is evaluated on the extent to which each method (still targeting the poorest 29%) provides benefits to individuals consuming less than the international poverty line of USD $1.90 per person per day (76% of observations in the 2020 phone survey dataset and 57% of observations in the 2018–2019 field survey). Spearman correlation and AUC are not reported here as they do not depend on the classification threshold, and are thus identical to the values reported in Table 1.