Participants for the cross-sectional survey were sampled using multi-stage sampling; first sampling wards within selected study districts, then by randomly selecting streets, then households within those streets (random selection of the first house, systematic thereafter). Individual structured interviews were conducted in person by trained interviewers, with the aid of tablet devices (questionnaires programmed in ODK). 599 women and 600 men participated, whose data are included in the datasets. Data cleaning steps included checks for duplicate IDs, missing data, verifying age of respondent, consistency checks across related variables, and checks for correct application of questionnaire filters. The dataset was anonymised with each record assigned a unique numerical ID.