OUR PARTNERS
This programmatic option describes the conduct of data analysis on types, intensity and periods of electoral violence.
The nature and history of electoral violence are key considerations in assessing the prospect of potential future election-related violence, and they inform the launch of prevention initiatives. The use of open-source data can supplement official national sources including State-led and/or NGO-led early warning efforts to assist in analysing the type, intensity and periods of electoral violence. Several resources now exist that focus on social and political incidents of violence.
The collection, classification and collation of data has largely been undertaken by think tanks or academic institutions that are able to maintain a sustained commitment to the exercise. These systems often rely on media reporting as a principal source of data with secondary efforts to authenticate incidents. A traditional challenge of such systems has been the capacity to capture and classify events in a timely manner, verify their authenticity and to authoritatively identify perpetrators. However, the use of new technologies is closing the gap between the occurrence and capture of data, as demonstrated by the near real-time Armed Conflict Location and Event Data (ACLED) resource.
In many countries, certain challenges remain around incident reporting on electoral violence, such as:
The primary objectives of conducting an analysis of electoral violence data are:
In determining the use of external data sources, the relevant codebook that describes the classification system should be consulted to ensure that the classification of incidents is clearly understood and interpreted to acquire the relevant learnings from the studied data.
Where data on electoral violence can be found, certain caution should be taken when determining their credibility and trustworthiness, considering issues such as impartiality, independence and the applied methodology. Nowadays, several resources provide dynamic dashboards to allow for interactive analysis. Analysing the data requires some level of proficiency but does not require specialized software. Most systems will allow data to be downloaded and analysed using Microsoft Excel.
The implementation of electoral violence data analysis is best led by a combination of key actors:
These actors, working together, ensure accurate, timely analysis and help develop effective strategies to prevent electoral violence.
To ensure an analysis of electoral violence data is context-specific, it is important to start by understanding the unique political, social, economic and historical factors influencing the situation. This involves identifying key variables such as political power dynamics, ethnic or religious conflicts, historical tensions, local governance structures and cultural attitudes towards political competition. A combination of data sources should be used, with local insights gathered through interviews, fieldwork or focus groups to complement quantitative data. Additionally, data should be broken down by factors like region, gender or political group to highlight how violence affects different segments of the population.
To maintain sensitivity, ethical practices must be followed, particularly when working with vulnerable groups, by ensuring confidentiality and recognizing the risks for those providing sensitive information. It is essential to include the perspectives of marginalized communities and collaborate with local experts to ensure the findings reflect local realities accurately. When presenting the analysis, it should be framed carefully to avoid causing harm or escalating tensions.
To involve youth in the analysis of electoral violence data, incorporate the following:
To ensure gender-sensitive and inclusive analysis of electoral violence data, it is important to disaggregate the data by gender and create indicators that reflect the unique experiences of women, men and non-binary individuals. Special attention should be given to gender-based political violence, harassment and intimidation, which may disproportionately affect women and marginalized groups. Including diverse gender perspectives in data collection, through interviews and focus groups, ensures that the analysis reflects the realities of different communities. Collaborating with gender experts and local organizations strengthens the analysis and helps design programmes that address specific risks and challenges. Ultimately, findings should be presented using inclusive language that promotes gender equality and avoids reinforcing bias or stereotypes.
The communication of this activity can be undertaken through the release of reports, newsletters, social media, videos or other modalities. When doing so, it is important to clearly outline the source of the information, the methodology applied and the findings observed from the conduct of the analysis of electoral violence data.
The coordination of actions with other international and/or national entities may enrich the collection, analysis and interpretation of electoral violence data, as well as further provide legitimacy to the findings of the process. Furthermore, collaboration with third-party actors may also bolster the visibility and reach of the exercise with regards to citizens, experts and other organizations that may be interested in the findings.
To ensure the sustainability of an exercise concerning the analysis of electoral violence data, recurrent updates and amendments will need to be introduced to adapt it to ongoing and upcoming processes and parameters. If and when the source or methodologies used change, it is important to present it in a transparent and clear manner to allow for a better interpretation of the historic data analysis.
Conducting an electoral violence data analysis may need to take into account certain costs linked to this exercise:
Analysing electoral violence data is a complex task that involves navigating various challenges related to data quality, context and methodology. Several relevant limitations and challenges can be found when conducting this exercise, such as:
Armed Conflict Location and Event Data (ACLED)
ACLED is a conflict collection, analysis and crisis mapping project with coverage across countries in Africa (1997 onward), South and South-East Asia (2010 or 2016 onward) and the Middle East (2016 onward). The system provides an interactive dashboard as well as downloadable data sets. The information is updated in near real time based on researchers primarily extracting information from secondary source information. ACLED collects the dates, types of actor (eight coded), types of violence (nine coded), locations and fatalities associated with reported political violence and protest events.
Uppsala Conflict Data Programme (UCDP)
Based at Uppsala University in Sweden, UCDP is a leading research initiative that collects and disseminates data on global armed conflicts and organized violence. Established in the 1970s, UCDP provides comprehensive, publicly accessible data sets on various types of conflicts, including State-based conflicts, non-State violence and violence against civilians. Its data are widely used by researchers, policymakers and organizations to analyse conflict dynamics, track trends over time and support efforts in conflict resolution and peacebuilding. UCDP is a key resource for understanding global patterns of violence, including in the context of elections and political instability.
Global Terrorism Database (GTD)
The GTD provides data from around the world on terrorist attacks in the period from 1970 to 2017. The events include domestic as well as transnational and international incidents providing information on the date, location, weapons used, nature of the target, casualties, and the group or individual responsible (when available). The GTD definition of terrorist acts is broad, encompassing a range of political motivations. The data are open source and provide records of more than 180,000 incidents worldwide with between 45 and 120 variables recorded for each instance.
National Elections across Democracy and Autocracy (NELDA)
The NELDA data set provides extensive data about individual rounds of elections through 53 variables across 200 countries for the period 1945 to 2012. Specific to issues of election-related violence variables are numbers 29, 30 and 31 (riots and protests after the election); 33 (significant violence involving civilian deaths before, during or after the election); 41 (leader replaced due to widespread protests); and 42 (a coup prevented the elected leader from taking office).
Social Conflict Analysis Database (SCAD)
The SCAD data covers all of Africa, Mexico, Central America and the Caribbean over the period 1990 to 2016. Each recorded event – more than 20,000 – provides information on the location, timing and magnitude of social events including actors, targets, issues of contention and governmental response. The events recorded do not require the presence of armed conflict but focus on capturing social conflict events including inter alia protests, riots, strikes, inter-communal conflict and government violence against civilians. The data provided are researched by AP and AFP news wires.
For more informations contact : [email protected]
follow us on Twitter