The CPOST Suicide Attack Database (CPOST-SAD) contains data on the universe of suicide attacks. For each attack, the database includes information about the geographic location, target classification, and weapon(s) used, as well as systematic information on the demographic and general biographical characteristics of suicide attackers. Here we briefly explain the criteria for including and coding attacks.

What constitutes a suicide attack? 

CPOST defines a suicide attack as an attack in which an attacker kills himself or herself in a deliberate attempt to kill others.  CPOST includes only suicide attacks perpetrated by non-state actors; attacks authorized by national governments are not included.  The classic example is a suicide bomber detonating an explosive vest (a “belt bomb”) or explosives in a vehicle the bomber is driving (a “suicide car bomb”).  The critical criteria is suicide: the attacker must kill him or herself, even if no one but the attacker dies in the attack.  The CPOST-SAD does not include (1) failed suicide attacks where explosives do not detonate or are detonated by someone other than the attacker (e.g. the explosives were set off by a gunshot from police); or (2) “suicide missions,” where the attacker expects to be killed while killing others, but does not directly kill himself or herself. To be included in the database, suicide must be verified by two independent sources. Including only verified completed suicide attacks -- and making the sources used to verify each attack available to the public -- distinguishes the CPOST-SAD from other databses that collect suicide attacks.

Must suicide attacks qualify as “terrorism” to be included in the database? 

Suicide attacks are often associated with “terrorism” in the media, policy, and academia.  The boundaries of what counts as “terrorism,” however, is notoriously difficult to define. Because CPOST focuses collection on a specific mode of attack (suicide), CPOST can afford to be agnostic on whether a given suicide attack qualifies as “terrorist” or might better be described as some other form of violence (for example, a guerilla attack or an attack with no obvious political motivation).  Beyond the non-state qualification, any attack meeting the definition of suicide attack is included, conditional upon verification by two independent sources.

What sources does CPOST use to identify and confirm attacks?

To identify and verify suicide attacks, CPOST uses open-source news websites and archives such as Lexis Nexis and  These are the primary sources for data on attacks and attack variables such as targets, weapons, and casualties. In addition, CPOST uses militant group websites, martyr videos, and social media accounts (e.g. Twitter) to locate claims for attacks and attacker biographical data not reported in the media.

How does CPOST code target types? 

At the most basic level, CPOST categorizes targets as Security, Political and Civilian. Security targets include the military, police, or intelligence forces and their support structures (e.g., bases, supply lines, civilian employees), as well as rival militias and militant groups. Political targets include politicians and public officials, elected and unelected, foreign and domestic, at all levels of government.  It also includes their offices and supporters. A target is coded as Civilian when the intended victims have no direct affiliation with either security or political structures.

Suicide attacks against security and political targets often kill many civilians. In such cases CPOST codes the intended target – that is, the target most likely to have been selected by the attacker based on reports. For example, if a suicide bomber attacks a restaurant where a high-ranking government official was dining, the attack is assumed to be targeting the government official, not the restaurant, and is coded Political.  If no information on a likely target is present, the target is assumed to be civilians dining at the restaurant and the attack is coded as Civilian.

To return to the database, click here.