Sampling Procedure
The sample for the Malawi RaMMPS was generated via (screened) Random Digit Dialing (RDD) without replacement. Using the mobile phone numbering structure in Malawi, a set of random numbers was generated by Sample Solutions (https://sample.solutions/), and verified against the Home Location Register (HLR), which is a database of registered (including pre-paid) numbers on the Global System for Mobile Communications (GSM) network. Enumerators conducted Computer Assisted Telephone Interviews (CATI) with active strata monitoring. Strata were a-priori defined in terms of broad age groups (18-49 and 50-64), sex, region (North, Central, and South), and type of place of residence (urban/rural), and the quotas for each stratum were derived from the 2018 Malawi Population and Housing Census. Once a stratum was filled, respondents with these attributes were no longer eligible to participate in the study. Fieldwork was divided into four blocks of 4 to 5 months each, and quotas were re-set at the beginning of each fieldwork block. Quotas for some of the strata were difficult to fill, and therefore an Interactive Voice Response Survey (IVR) was also used to identify respondents living in rural areas.