During periods of armed conflict, there are structures, actors, and processes at a number of levels that affect the likelihood of violence against civilians.
In direct violence perpetrators act face-to-face with the victims using small weapons such as machetes or rifles.
[12] Targets may also be chosen selectively, identifying specific individuals who are seen as opposing a political group or aiding its opponents.
[6] The decade following the dissolution of the Soviet Union was marked by a decline in worldwide battle deaths and the number of armed conflicts in the world.
However, foreign actors that are democracies or have strong human rights lobbies are less likely to support groups that engage in violence against civilians.
Perceptions of threat due to external military intervention may lead to increases in governmental mass killing of civilians[22] and violence against domestic out-groups.
[24] Relationships between government (at various levels), armed groups and domestic populations affect violence against civilians.
Rebel groups that control areas inhabited by nonconstituents are more likely to use violence to obtain resources and cooperation.
[6] Ideology may strongly influence the ways in which governments and rebels define their constituencies, affecting patterns of violence.
[25][26][6][9] Violence against civilians may vary over space and time with the extent to which military forces are contesting a territory.
[27] Opinions differ widely on whether there is a relationship between the relative military capacity of a government or rebel group and the likelihood that it will engage in patterns of violence against civilians.
[28] At the organizational level, researchers have examined the dynamics and ideology of armed groups: how they recruit and train their members, how organizational norms about the use of violence against civilians are established and maintained, and the role of group leaders and political ideology in shaping organizations and behavior.
[6] Jeremy Weinstein has argued that armed groups develop certain organizational structures and characteristics as a result of their available resources.
According to this view, organizations that depend on external resources are predicted to attract low-commitment members, and have trouble controlling their use of violence against civilians.
[6] On an individual level, people may be influenced to participate in armed conflicts due to economic motivations or incentive structures.
Research in this area often views violence against civilians as a by-product of economic processes such as competition for resources.
[6] While research suggests that emotions such as fear affect the polarization of attitudes, material and structural opportunities are important mediators of the expression of violence.
[34] At the individual level, researchers are examining the category of “civilian" in greater detail, to better understand the use of violence against different types of noncombatants.
[38][39] A relatively new area of research asks how individuals, groups, communities and domestic and international audiences respond to violence against civilians.