LRA Crisis Tracker Methodology

1. Data Collection

Report sourcing:

  • HF radio operators in the Democratic Republic of Congo and Central African Republic
  • Civilians report activity to HF radio tower operators
  • Over 70 HF radio operators call the Dungu and Obo hubs twice daily to report armed group activity
  • Activity is entered into a spreadsheet and then sent to data coders
  • UN and NGO reports
  • News and media outlets
  • Civil society contacts in local communities
  • Government sources
  • Field research conducted by Resolve and Invisible Children staff

Sourcing coverage: LRA Crisis Tracker Database team members make every effort to obtain data from all LRA-affected regions. Due to the remote nature of LRA-affected areas, the sourcing infrastructure available to project administrators is uneven across the geographic area of concern, and data included in the Database is often of better quality in areas with higher NGO and news agency traffic. The LRA Crisis Tracker Database does not claim to be a comprehensive record of all LRA or related incidents in the region, but team members make every effort to fill in areas where the data may not be easily accessible.

Note: The majority of the information gathering systems are located in Congo, leading to a disproportional amount of LRA reports from DRC. In upcoming months both Invisible Children and CRS, funded by USAID, will expand information gathering systems in CAR, hoping to improve access to information in the region.

1. Data Entry

Database entry: Reports are divided between a team of coders from both Invisible Children and Resolve. Coders determine if the source is reliable or unreliable (See section 4.2.B of the Codebook, Determining the Reliability of a Source). Before an incident is reported, the coder reads through other incidents in the same time range and checks for duplicates.

Verification rating: After an incident is categorized, each incident is given a Verification Rating, which rates the team's confidence in the details of the reported data. Each incident is given a rating of "1" through "5," with "1" being the most unreliable and "5" being very reliable. The rating is based on the trustworthiness of the source, confidence in the identity of the actors involved in the incident, and the degree of detail given in the source report. A verification rating of "2" through "5" is considered adequately verified to be reported publicly, and therefore is included in statistics and analysis (Codebook section 4.2A).

LRA Actor Verification rating: To distinguish between LRA and other armed group attacks, the Crisis Tracker Codebook has a list of LRA Indicators and Non-LRA Indicators. If after reviewing the indicators and other available evidence the data coder determines that the perpetrator of an attack was likely the LRA, the incident is given an LRA Actor Verification rating, 'Low,' 'Medium,' or 'High,' to measure the likelihood of the perpetrator being LRA. (Codebook section 4.2C.) If after reviewing an incident the coder determines that the LRA was not the perpetrator and the perpetrator is unknown, Actor 1 is marked as 'Armed Group' and the incident is not mapped.

3. Data Review

Initial review: Each report is reviewed by a second data coder to catch human errors and duplicate reports. Coders look for incidents that are alike in detail, and have a relatively close time frame and location. These incidents are then investigated to ensure that they are not duplicate reports.

Expert review: IC and Resolve staff with field experience review sensitive incidents immediately and review all incidents every three months. Should this staff member feel an incident was misreported, the incident is corrected and potentially unmapped. External LRA and regional experts are consulted as necessary.

4. Data Mapping & Sharing

Data mapping: After an incident is entered and approved to be mapped, it appears on the LRA Crisis Tracker website. Only incidents involving the LRA or persons formerly abducted by the LRA and given a Verification rating of '2' or higher are mapped.

Data sensitivity: Sensitive information such as specific sources, names, information on security forces, and personal information about minors is not shared publicly.

Data sharing: Data is regularly sent to UN agencies and humanitarian practitioners for comparison and collaboration.

5. Data Revamp
As the database grows and policies are updated to reflect best practices, data coders revisit and "revamp" the data when needed.

With the establishment of the HF Radio Network and expanded reporting mechanisms in the region, incident reporting has become more detailed and the database has been adapted to reflect this. Fields including information on age and gender of victims, and goods looted have been added since the beginning of the database. Coders periodically revisit all incidents and reports to include the new details and fields.

6. Data Analysis & Reporting

Crisis Tracker staff analyze data for trends and patterns in LRA activity. For instance, coders look for trends in the age and gender of abducted persons, net recruitment (total abductions- total returnees), and increases in a certain type of attack. Coders also look for new traits and patterns in LRA activity.

Specific areas and provinces are also analyzed for increases or decreases in number and type of attack.

After analysis has been completed and reviewed, it is reported in various Crisis Tracker reports.



An incident is considered an "attack" in the Brief if LRA activity results in one of the following human rights violations on one or more civilians: violence resulting in death or injury, sexual or gender based violence, abduction, or looting. For detailed definitions of these human rights abuses, please refer to section 4.5 of the LRA Crisis Tracker Map Methodology and Database Codebook v1.6.


An incident is regarded as a "killing" if there is a violent act that results in the death of an individual who is not known to be associated with an armed group or security force. Civilian deaths resulting from injuries sustained from an attack are considered a "killing." Also, if a civilian is killed while in LRA captivity, it is considered a "killing" if it occurs within one week of the initial abduction. For a detailed explanation of incidents that are categorized as a "killing," please refer to section 4.5.1 of the LRA Crisis Tracker Map Methodology and Database Codebook v1.6.


An incident is regarded as an "abduction" if it involves one or more persons taken hostage against their will by the LRA for any period of time, including civilians who are abducted and released or escape in the same day. A short-term abduction is considered any abduction that is 72 hours or less in duration. This does not necessarily mean that abductions that are not short-term are long-term as there may not be a report of the abducted person's return. For a detailed explanation of incidents categorized as "abductions" or "short-term abductions," please refer to section 4.5.2 of the LRA Crisis Tracker Map Methodology and Database Codebook v1.6.


A "returnee" is considered anyone who escapes, is released, is rescued, or defects from LRA captivity. It also includes all LRA members who are captured. For a detailed explanation of data relating to returnees, please refer to section 4.5.2 of the LRA Crisis Tracker Map Methodology and Database Codebook v1.6.