Mobile Myanmar: The Impact of Social Media on Young People Living in Conflict-Affected Regions of Myanmar When asked if they trust in news seen on social media, the most common response was ‘sometimes’ (42.2%), followed by often/usually (33.2%). The distribution of responses by sub-sample is shown in Figure 6. When responses were converted to an ordinal 1-5 point scale, Kruskal-Wallis tests revealed that on average, IDP respondents had a significantly higher level of trust in news seen on social media compared to urban and rural respondents (p < .001), and males a significantly higher level of trust compared to females (p < .05). Figure 6: How often trust news from social media 70% 60% 50% Kayah Urban 40% Kayah Rural 30% Rakhine Urban 20% Muslim IDP 10% 0% Almost never Seldomly Sometimes Often Always Most of the overall sample indicated that they check the accuracy of news read on social media (65.1% overall), with rates of checking highest among the IDP (89.5%) and male (77.4%) sub-samples. Of those who did check the accuracy of news read on social media, the most common methods overall were to ask friends or family (60.4%), check other Facebook pages (59.7%), and check other news sources (50.0%). These three were the top-ranked methods across all demographics and sub-samples (see Figure 7). However, males and the Rakhine State sub-samples were much more likely to select more than one method, and so as per previous ‘select all that apply’ questions, had higher rates for each option compared to females and the Kayah State sub-samples respectively. Differences worth noting were that IDP respondents were most likely speak to community leaders or check the Facebook page of local reporters, while rural respondents were least likely to check other news sources. Figure 7: Checking accuracy of news on social media 80% 70% 60% Kayah Urban 50% Kayah Rural 40% Rakhine Urban 30% Muslim IDP 20% 10% 0% Check Use Google Ask friends Check Speak to Check with other news or family another community local sources F'book page leader reporter F'book page 91 Other

Select target paragraph3