-p-1080.jpg)
.jpg)
In a world where violence against women remains a critical human rights issue, the need for comprehensive and reliable data to understand and combat femicide is more pressing than ever. Femicide, the gender-motivated killing of women, not only devastates families and communities but also reflects the deeply entrenched gender inequalities that plague societies worldwide. Recognizing the significance of this issue, the Odipodev and Africa Data Hub team embarked on an extensive project to compile a detailed database of femicide cases in Kenya, aiming to shed light on the prevalence and nature of such crimes within the country. This blog post outlines the rigorous methodology we employed to gather data on femicide cases reported in Kenyan news media from January 2016 to December 2023, adhering to the criteria established by the United Nations Office on Drugs and Crime (UNODC).
The methodology
1. Determining our approach
This data was obtained through a systematic and meticulous search for femicide cases reported in Kenyan news media between January 2016 to December 2023; it is limited to femicide cases only reported in Kenya’s news websites and reported in English. We decided to use news stories as our primary source of information as it is near impossible to obtain this kind of data from other institutions, such as the local police. News stories often also provide additional context that official statistics may not share, including circumstances around the death and the relationship between victim and perpetrator.
2. Defining our scope
The curation was done by adhering to UNODC criteria for identifying gender-motivated murder of women (femicide).
The UNODC’s criteria includes:
- The homicide victim had a previous record of physical, sexual, or psychological violence/harassment perpetrated by the author of the killing
- The homicide victim was a victim of forms of illegal exploitation, for example, in relation to trafficking in persons, forced labour or slavery
- The homicide victim was in a situation where she was abducted or illegally deprived of her liberty
- The victim was working in the sex industry
- Sexual violence against the victim was committed before and/or after the killing
- The killing was accompanied by mutilation of the body of the victim;
- The body of the victim was disposed of in a public space
- The killing of the woman or girl constituted a gender-based hate crime, i.e. she
3. Identifying the search terms
To find those articles covering the cases mentioned above, we first had to know the search terms that would be used to surface relevant articles, and explore which words were consistently used while reporting femicide. By reading through dozens of articles that reported on femicide, we identified a consistent set of words that are used to describe victims (woman, girl, wife, girlfriend) and the act of murder (murder, murdered, kill, killed).
We experimented with different strings of words and ultimately used the syntax below to retrieve all news stories from major Kenyan news outlets with the combination of the two sets of keywords:
- Words used to describe victim - (woman, girl, wife, girlfriend)
- Words used to describe murder - (kills, killed, kill, murdered, murder, death, died, dead).
Our exact syntax (including the newsrooms we wanted to search) is below:
woman OR girl OR wife OR girlfriend AND kills OR killed OR kill OR murdered OR murder OR death OR died OR dead site:nation.africa OR site:citizen.digital OR site:www.the-star.co.ke OR site:standardmedia.co.ke OR site:www.k24tv.co.ke OR site:www.capitalfm.co.ke OR site:pd.co.ke
4. Manually Curate and Verify Data
Using Advanced Google Search, we applied the string above and any article combining any of the words from the first set (victim) and the second (murder act) appeared in the results. In an effort to ensure that no cases were missed, we used the wide range of search words above. The results from the search yielded a huge chunk of irrelevant articles - over 11 000 articles! We manually sifted through this list article by article to find the murders that qualify as femicide as per UNODC guidelines. This process of filtering reduced the number of stories to 1000. From the filtered cases, we then read each article's text to pull out key details such as the victim's name, victim's age, date of murder, location, and scene of murder, relationship with the suspect as well as additional circumstances around the murder. Where possible, we also looked up any additional details about the victim and perpetrator that one story might not have captured. In some cases, we could access court records or use other stories to provide additional details. In many other instances, however, this additional information was not available.
5. Organise and Catalog Data
The output was a spreadsheet file where all these details were captured along with the story details such as date of publishing, headline, publisher and full article text. The result of that work is the database used in this list and is the basis of our analysis.
We might have missed some cases, but we are continually updating the database. If you know of any case we missed please provide us with the details using this link.
The creation of this femicide database marks a significant step towards understanding the scope and specifics of gender-based violence in Kenya. By adhering to the UNODC's stringent criteria and by employing diligent data collection techniques, our methodology ensures that each case included in the database is a true representation of the dire issue at hand.
This database not only serves as a resource for researchers and policymakers but also acts as a call to action for all stakeholders involved in the fight against gender-based violence. We acknowledge that our database, while comprehensive, may not capture every instance of femicide. Therefore, we invite the public to contribute by providing information on any cases we might have missed, ensuring that our database remains up-to-date and reflective of the ongoing struggle against femicide in Kenya (use this link). Together, with accurate data and collective effort, we can pave the way for meaningful change and work towards a future where women can live free from the fear of gender-based violence.
