Questions people ask

Where do the events come from, and how are they checked?

MCARI's record is built from open-source reporting and from its own network of local sources. Reporting published by Myanmar outlets is monitored continuously.

Where several outlets report the same event independently, their accounts are read together and the agreement between them is the check. Where an event rests on one or two reports alone, it is put to people in the place concerned: residents of that township, or correspondents working there. An event that local sources cannot confirm does not enter the dataset. A claim made by a party to the conflict about its own operations is treated as a single-sourced report and checked the same way.

Why do MCARI's figures differ from other counts?

Counts differ with the rules behind them. MCARI records an event only when it can be confirmed (question 1). Where sources give different figures for the same event, MCARI takes the lowest figure any source gives, the one figure every source supports, and does not average them. Nothing is estimated, interpolated or modelled.

MCARI's figures are therefore often lower than counts that include estimates or unconfirmed reports. The rules are set out in full in the MCARI Methodology and Codebook.

Are the death figures exact?

No. They are minimums. Where sources disagree, the lowest figure is taken. Where a source reports deaths without giving a number, a fixed conventional figure is recorded; where it gives a quantity in words, such as "hundreds", the least those words allow. The real figure is higher than the recorded one more often than it is lower, and by an unknown margin.

Where a death is reported without saying whose it was, it is counted in the total but not attributed to a side.

Does the dataset include everything that happened?

No. MCARI records what reporting establishes. Where reporting is thin, the record is thin: areas closed to journalists, periods of communications blackout and places far from any outlet's reach produce fewer recorded events, not because less happened there, but because less of it was reported. A zero means nothing was recorded. It does not mean nothing happened.

There is no field for the injured, and displacement counts recorded displacement events, not distinct people.

What does "Unidentified Armed Group" mean?

It is used where reporting shows that an armed group acted but does not establish which one. This is common, and it is a property of the reporting rather than a gap in the coding. MCARI does not guess a perpetrator: where a source only hedges at a name, the event is recorded as Unidentified Armed Group.

Why do figures on this site differ from a published issue?

MCARI's dataset is a living record. A row is corrected when better information reaches it, and corrections are not held back to preserve a published figure.

Each issue is coded from a frozen edition, which it names, and every figure in it stays reproducible from that edition. This site shows the current data release, which includes every correction made since. Where the two differ, the issue's figure is correct for its edition, and the site's figure is correct for the current release.

How often is the data updated?

MCARI's dataset is updated monthly. Each update is published on this site as a new numbered data release. "What changed in each release" and the checksums are on the Methodology & Data page.

Why is an event or a township missing from the map?

A township map can show only events placed to a township. Where a source gives only a district, a region or a nationwide measure, the event is recorded but cannot be drawn, and the number of such events is stated under the map.

One township, Ho Tawng, is not drawn by the boundary file the map uses; its events are named with their count in the map note rather than dropped.

Can I download the full dataset?

The full dataset is not published. Township maps, charts and every issue of The Myanmar Conflict Monitor are open to all, and the figures on each Places page can be downloaded as a CSV file.

Row-level data is supplied on request; a request should state the period, the geographic scope and the purpose it is needed for. Write to info@mcari.org.

How do I cite MCARI, and can I reuse the charts?

Every issue of The Myanmar Conflict Monitor is published under the Creative Commons Attribution 4.0 International licence (CC BY 4.0): it may be copied and reused with credit to MCARI. Charts and figures on this site may also be reused under the same licence, with credit to MCARI.

Figures from this site are cited as: MCARI, Myanmar Conflict Data, [release], Myanmar Conflict Analysis Research Institute, mcari.org, accessed [date]. Other forms of citation are on the Methodology & Data page.

Why do I need an account to download issue PDFs?

The web edition of every issue is open to all. An account is needed for issue PDFs, the event-level map and data requests. MCARI stores only your email, organisation and purpose, and a record of the issue PDFs you download. No password is created: you sign in with a link sent to your email. The Privacy Policy sets out what is kept and why.

I found an error. How do I report it?

Write to info@mcari.org, with the event, the place and the date, and what you believe is wrong. Corrections found after an issue is published are recorded in the Corrections log and carried into the next edition.

Is MCARI independent?

MCARI is not affiliated with any government, armed group, or political organization. Its findings are not adjusted to reflect the interests of funders or partners. The rules behind every figure are published in the MCARI Methodology and Codebook, with their reasoning, so that others can assess them and, where necessary, challenge them.

How does MCARI protect its sources?

The people who confirm events for MCARI are not named, and nothing MCARI publishes identifies them. Each event carries a note recording what local sources established. Notes are never published or supplied, because releasing them would expose the people who did that work. Members of the field team are not publicly identified.

Is a cold spot the opposite of a hot spot? Why are Naypyitaw and Yangon not all cold spots?

Not quite. A hot spot or a cold spot is a group of neighbouring townships, not one township on its own. Each township is compared, together with its neighbours, with the national average, and the map marks it only where the difference is clearly larger than chance would give. Most townships are neither, and are shown in grey.

A cold spot is a group of neighbouring townships where the count is clearly below the national average. It does not mean the area is safe or peaceful: it means fewer events were recorded there.

In 2025, nine Yangon townships were cold spots for deaths: Bahan, Dagon, Lanmadaw, Mayangone, Mingalartaungnyunt, Pabedan, Tamwe, Thingangyun and Twantay. The other Yangon townships were grey, because they or their neighbours recorded more.

No Naypyitaw township was a cold spot, although several recorded no death in 2025, because their neighbours outside Naypyitaw were active: Yedashe recorded 203 deaths, Thandaunggyi 89 and Pekon 88.

Hover over a township on the hot spot map, or find it with the search box, to see its own count and its busiest neighbours.