Mobile Myanmar: The Impact of Social Media on Young People Living in Conflict-Affected Regions of Myanmar At the Participant Level, purposive sampling was used to select FGD and KIIs participants to ensure that the overall sample was:  Gender balanced  Religiously diverse  Ethnically diverse  Educationally diverse  Age diverse Recruitment of participants was achieved through snowball sampling (using existing SC-Myanmar and Research Associate networks), which had the benefit of being logistically straightforward. However, it should be noted that this method potentially introduced additional confounding factors into to the analysis regarding how offline social networks might influence online use. Participants for the FGDs were identified and recruited by SC-Myanmar staff, and in the case of Rakhine, by the Research Associates’ networks. In Loikaw, young people were recruited by field-based education staff while in Sittwe, young people were recruited by Education in Emergencies (EiE) and Child Protection in Emergencies (CPie) staff. These urban and village-based participants were recruited from SC-Myanmar target areas and are direct or indirect beneficiaries of SC-Myanmar programming. Muslim IDP recruitment was facilitated by Rohingya camp-based staff, such as camp-based education liaisons and program assistants, who interact with their communities on a daily basis. Recruitment of Rakhine, Chin, and Maramagyi young people was facilitated by the Research Associates’ networks in Sittwe, with many of these participants identified by local Rakhine civil society organizations (CSOs). KII participants were recruited by SCI-Myanmar staff, using snowball sampling from FGD participants, and in some cases by convenience sampling in public spaces. Tables 1 and 2 present a summary of the demographic profile of FGD groups and KII participants across locations. 30