A Regional Overview (RO) highlights significant trends and events that ACLED records during the previous coding week (Saturday – Friday). It is a short introduction and, as such, should not provide an exhaustive account of everything that happened but rather highlight key/novel trends and events within the region. Authors should place particular emphasis on events/trends involving violence (both organized and spontaneous), as well as demonstration movements that challenge state power/legitimacy.
An RO has two core components – an introductory paragraph that outlines and links core trends; and a body of paragraphs broken up according to countries and/or specific trends. Unlike other ACLED publications, an RO does not require a conclusion.
For examples, see: https://acleddata.com/regional-overviews/
An RO should not exceed two pages in length.
Contextualization – authors must provide the context for WHY an event is included in the RO. How does an event/trend situate within the wider landscape? Is it a continuation of a pre-existing and worrying trend? Or, is it a new shift – if yes, how has it shifted and why is it significant?
Background information/supporting evidence – background information can be useful to highlight the importance of a particular trend or event. When making an argument or claim that ACLED data do not support, authors must include referenced evidence to support such arguments/claims. In particular, authors are encouraged to link the text to previously published ACLED analyses where appropriate.
Early Warning (EW) outputs – specific references to weekly warnings provided by the ACLED Early Warning Research Hub. Weekly warnings and canned language for use in the RO should be downloaded from Weekly Early Warning File app from ACLED Apps. Authors should prioritize the inclusion of those flags that are marked as "High" priority outputs.
Pertinent non-ACLED events – if ACLED data do not include an important event, such as an election or key political announcement, this can and should be included in the RO. Authors should always provide references for such events.
Dates of important events – when including notable events, such as political announcements, elections, anniversaries, and assassinations, authors should include the specific date of the event within the text.
Short descriptors of included actors – when the nature of a particular actor is not immediately apparent, authors should include brief descriptors. For example, when referring to the New People’s Army (NPA) in the Philippines, an author could refer to the NPA carrying out a “communist insurgency.”
“Last week” – authors should regularly remind the reader that the RO is discussing the events of last week. For example, “police clashed with demonstrators across the country last week.”
Events falling outside the coding week – an RO should only include events occurring during the previous coding week (Saturday to Friday). Authors should leave events that arise after Friday for the following RO. In cases where a significant event occurs after Friday, authors can include a footnote acknowledging the event and reminding the reader that it will be included in the next RO.
Specific event numbers/statistics – as there are often delays in reporting events, specific percentages and event counts are likely to change in the week/s following any data release. Authors should, therefore, avoid using exact numbers in ROs. Authors can instead use vaguer language to describe trends, such as “demonstration activity decreased” or “the number of protest events more than doubled.”
Every country – it is not necessary to include trends from every country in an RO. An RO should highlight only the most significant/notable trends and events across the region.
Lists of events – ROs should not provide extensive reports of every event associated with highlighted trends. It is much more effective to identify one or two events within the data that typify the trends that you are highlighting.
Emotive language – avoid using emotive language (e.g., brutal torture) or language that attributes emotions to actors (e.g., angry demonstrators).
Unreferenced claims – do not include potentially controversial/political claims about a specific event/group without qualification. When including such claims, the author should always clearly identify the person/group that has made the claim and provide references.
RO drafts must be ready for initial review by end of day Tuesday each week.
Authors should save their ROs to a google doc in the Weekly Analysis Submissions folder in the appropriate weekly folder (e.g., August 2, 2021 – if the week’s folder isn’t there yet, create it). Within the weekly folder, authors must create a new regional subfolder each week (e.g., “Middle East”).
When the RO is ready for an initial review, tag the Analysis Coordinator in a message on the relevant regional analysis slack channel. Include a link to the doc.
Once the Analysis Coordinator has completed an initial review, they will tag you on slack.
Review the doc, accept all edits and address any comments. The doc must be clean -- no unresolved comments/edits -- before the next review.
If you have not already included EW outputs, these should be added.
Tag the Director of Research & Innovation in a new slack message for further review.
After the Director of Research & Innovation has completed their review, they will tag you on slack.
Review the doc, accept all edits and address any comments using the suggesting function.
Tag the Director of Research & Innovation in a new slack message for the second round of review.
Accept all final edits. If there are significant edits/comments in this round of review, the RO may require a further round of review.
Once the Director of Research & Innovation has provided final approval, tag the Senior Communications Manager and Communications Officer in a message on the analysis_final_links channel. This message must include a link to the final doc.
IMPORTANT: If you come across any proposed edits that you disagree with, please comment on the edits rather than directly rejecting them. Edits should only ever be accepted or discussed with the reviewer.