Methodological note

The group behind the Myanmar Social Media Monitoring Project remains anonymous to protect the identities of those involved. However, it is important to note that the analysts working on the project are native Burmese speakers currently residing outside of Myanmar. Their fluency in Burmese and understanding of the political and social contexts give them tacit knowledge about norms and idiosyncrasies of social media usage among Myanmar people. Simultaneously, the ongoing crisis has had and continues to have an impact on their daily lives. While being mindful that analysts may be biased as they find themselves in an insider position observing the situation as it unfolds, the project takes the utmost care to remain impartial and minimise potential bias.

Monitoring social media is a combination of quantitative and qualitative analyses. It was decided from the outset to focus on the most popular social media platforms such as Facebook, Twitter and TikTok and monitor developments there. To assist with this, the analysts applied Facebook’s CrowdTangle tool and set up an application programming interface (API) to monitor Twitter and TikTok.

Although there is a lot of data involved in monitoring social media, the process is not as automated as people might think as you have to select what to monitor. This means that in the instance of monitoring of military and pro-military accounts and groups, you have to know who they are. The project therefore started with a list of known supporters and spreaders of pro-military propaganda and has during the lifetime of the project expanded on this list in collaboration with other monitoring projects. This work is not without its own challenges as groups and accounts are often closed without warning (either by the administrators themselves or by social media platforms if the accounts violate community guidelines) and new ones are established daily. However, the project has monitored approximately 300 pro-military or military accounts on a regular basis and monitored trending hashtags on all three platforms in both English and Burmese to identify case studies and examples for analysis.

Different analytical methods have been used to produce the findings presented in this report. To analyse the military’s approach and toolboxes, evidence and case examples have been gathered over the course of six months to identify patterns and corroborate findings. A similar method was applied to analyse patterns of communication within the protest movement. For the network analyses, the project has looked at linkages between the monitored accounts and groups through the use of referrals and handles in posts. While these approaches have their limitations – not least due to necessary privacy barriers – they have proven useful in determining overall trends and emerging issues in the Myanmar social media sphere.