CIV_VIOLENCE consists of a set of 22 separate variables, each covering one year from 1997 to 2018. The variables report the number of days, within the specified year, in which violence against civilians occurred within a 10-kilometer circular buffer around each DHS cluster location. By adding CIV_VIOLENCE to your data cart, you are in effect adding 22 separate variables (CIV_VIOLENCE_1997 to CIV_VIOLENCE_2018), one variable for each year. The 2018 data in IPUMS-DHS end on September 15, 2018.
Data for CIV_VIOLENCE are obtained from the Armed Conflict Location and Event Data Project ACLED [URL omitted from DDI.], downloaded on September 15, 2018. The Armed Conflict Location and Event Data Project (ACLED) defines violence against civilians as deliberate acts perpetrated by an organized political group (such as rebel, militia, or government forces) against unarmed non-combatants who are unable to defend themselves or engage in violence. Such conflicts harm or kill civilians, through events resulting in significant harm (e.g., bombing, shooting, torture, rape, mutilation) or accosting victims (kidnapping and disappearances); however, no fatalities are required for inclusion of the event. Events in which only property damage occurs or people are relocated without physical harm do not count as incidents of violence against civilians.
ACLED [URL omitted from DDI.] data provide differing levels of "geo-precision" or locational accuracy. IPUMS-DHS includes all levels of locational accuracy in its variable CIV_VIOLENCE. Please refer to ACLED [URL omitted from DDI.] for additional details on their procedures, methods of data collection, and data revision.
CIV_VIOLENCE, like the other IPUMS-DHS contextual variables, is calculated and based on sources external to the DHS. For more information on IPUMS-DHS contextual variables and how they are constructed, refer to the documentation [URL omitted from DDI.] on these variables.