Geographies of missing data: Spatializing counterdata production against feminicide

Abstract

Feminicide is the gender-related killing of cisgender and transgender women and girls. It reflects patriarchal and racialized systems of oppression and reveals how territories and socio-economic landscapes configure everyday gender-related violence. In recent decades, many grassroots data production initiatives have emerged with the aim of monitoring this extreme but invisibilized phenomenon. We bridge scholarship in feminist and information geographies with data feminism to examine the ways in which space, broadly defined, shapes the counterdata production strategies of feminicide data activists. Drawing on a qualitative study of 33 monitoring efforts led by civil society organizations across 15 countries, primarily in Latin America, we provide a conceptual framework for examining the spatial dimensions of data activism. We show how there are striking transnational patterns related to where feminicide goes unrecorded, resulting in geographies of missing data. In response to these omissions, activists deploy multiple spatialized strategies to make these geographies visible, to situate and contextualize each case of feminicide, to reclaim databases as spaces for memory and witnessing, and to build transnational networks of solidarity. In this sense, we argue that data activism about feminicide constitutes a space of resistance and resignification of everyday forms of gender-related violence.

Introduction

In a family home. In a desert. In a public square. On the side of a busy road. In a rural area. These are all spaces where women have been murdered or their bodies found, in cases of feminicide. They are also spaces where families and activists have taken action to denounce and refuse this violence. Feminicide constitutes the gender-related killing of cisgender and transgender women and girls. Violence against women, writes feminist geographer Damián Bernal (2009: 13), occurs in space-time: it is a conjunction of ideology, culture and gender relations as they manifest in a specific geography. In this article, we explore how space mediates the production of data about feminicide—by the state, the media and, crucially, by feminist and women activists.

Feminicide is an evolving feminist concept. Radford and Russell (1992) introduced the term “femicide” to refer to fatal forms of sexual violence and abuse. Latin American activists and scholars built on this to introduce the term feminicidio, which sought to attend to the state’s role in enabling violence against women through complicity or omission (Lagarde y de los Ríos, 2006; Monárrez Fragoso, 2002). Over time, and through activist struggles, femicide and feminicide have been increasingly taken up in legislation, national statistics and social mobilization. The terms typically denote murders in which a woman’s gender is part of the motivation for the crime, though legislation from different countries tends to vary in how crimes are typified. Here, we use the term “feminicide” based on the intersectional conceptualization of Fregoso and Bejarano, who brought Latin American formulations back into English as “feminicide” to describe how this violence is motivated not only by gender subordination but also by “the intersection of gender dynamics with the cruelties of racism and economic injustices in local as well as global contexts” (Fregoso and Bejarano, 2010: 5). This means most scholars and activists use an intersectional or transversal conception of feminicide, which understands that gender subordination is deeply interlinked with economic violence, colonial histories, and/or with racial oppression, among other structural forces.

Latin American feminist scholars have used a geographic lens to understand how this violence “unfolds in space and, in turn, reproduces spatial configurations that intensify women’s vulnerability, naturalize female death, and erase victims from public memory, enabling gendered geographies of violence” (Orozco, 2019). They have also used feminist methods for mapping and counter-mapping violence, both to understand its dimensions and to think through relations between body, geography, and power (Lan and Rocha, 2020). Feminicide is the ultimate expression of “how wars are fought through and over the body (and particular bodies)” of women (Faria et al., 2020: 1150). Beyond the loss of lives, these murders have deep ripple effects on the mental health and livelihoods of relatives and communities. They cause and perpetuate different forms of intergenerational trauma. They also leave family members more vulnerable to food, health and housing insecurities because women are so often providers of care and/or income (Fedina et al., 2022).

Since feminist and women activists first named and conceptualized violence against women as a result of gendered power differentials, they have been making the issue visible by compiling and publishing data. These encompass spreadsheets, maps, infographics, detailed reports, and the mobilization of data and art for public interventions. As we will discuss further, these data production efforts often—though not exclusively—emerge in response to voids left by the state and other institutions. Of course, official statistics about the problem do exist at different scales. According to United Nations’ estimates, around 89,000 women and girls were intentionally killed globally in 2022—the highest recorded number in 20 years. Nearly 55% were murdered by intimate partners or other family members (UNODC and UN Women, 2023). In Latin America and the Caribbean, every two hours, a woman is killed in violent incidents related to her gender (ONU CEPAL, 2016). In the United States, reports indicate that around three women are killed every day by their current or former male partners (McHugh, 2020). Widespread settler colonial violence against Indigenous women in North America (Sovereign Bodies, 2020) also demonstrates that gendered forms of violence that begin in times of “war” extend into and become normalized in times of so-called “peace” (Fregoso and Bejarano, 2010).

However, as staggering as existing statistics are, they are also known to be poorly documented and incomplete. Official government data on gender-related violence and feminicide, in particular, are often underreported, infrequently updated, contested or non-existent, despite the fact that families and civil society groups make repeated demands for such data to exist. This is due to multiple factors, as we will discuss further below, including legal complexities associated with the definition of the term, lack of infrastructure, poor training of state officials, and/or lack of political will. Drawing on other work, we refer to this phenomenon as missing data (Jungs de Almeida et al., 2023). In contrast with the missingness of official data, many grassroots initiatives have emerged in the past decades with the aim of documenting cases of feminicide and memorializing the lives of murdered women and girls. We consider much of this work to be counterdata production, a form of data activism that involves assembling counter-hegemonic case records, along with statistics, narratives, and interpretations of the phenomenon of feminicide (D’Ignazio et al., 2022; Milan, 2017).

Data activism on feminicide and fatal gender-related violence has intensified in the years following the #NiUnaMenos uprising, a wave of mass protests that started in Argentina in 2015 and spread across the whole continent (Chenou and Cepeda-Másmela, 2019; Revilla Blanco, 2019). Yet, it has a much longer history. It can be traced back to the 1895 publication of the Red Record by Ida B. Wells-Barnett (1895) in the US, the collection of data about the feminicides in Ciudad Juárez, Mexico and other Latin American countries in the 1990s (Benítez et al., 1999; Monárrez Fragoso et al., 2010), as well as work in North America to name and challenge the human rights crisis of Missing and Murdered Indigenous Women and Girls and Two Spirit people (MMIWG2S) (George et al., 2021; NWAC, 2009). In the last four years, we have cataloged more than 180 feminicide and fatal gender-related violence, monitoring efforts across the globe, led by individuals, data journalists, feminist collectives, academics and small non-profit organizations. Around a quarter of these projects started before 2004, but the majority have emerged in the last 20 years.

In this article, we examine how space, broadly defined, shapes both missing data and the counterdata production strategies of feminicide data activists. Much prior scholarship has demonstrated the spatial constitution and mediation of gender-related violence (Berlanga Gayón, 2015; Fuentes, 2023; Glockner et al., 2024; Monárrez Fragoso et al., 2010; Segato, 2004; Wright, 2009, 2013). We focus instead on how these connections are echoed in the spatial dimensions of data production (including non-production and counter-production) about feminicide. This study emerges from a larger research project called Data Against Feminicide that, since 2019, has aimed (1) to understand activists’ motivations and informatic practices through qualitative research, and (2) to co-design technologies with them. The analysis we present here draws specifically on our qualitative interviews with 33 feminicide monitoring efforts, based primarily in Latin America.

The contribution of this research is twofold. First, we put forth the notion of geographies of missing data as an analytical lens grounded on feminist principles for understanding the spatial politics of data production. This lens both introduces a spatial perspective into critical data studies and invites geographers and planners to engage with data production and information geographies as key objects of study. Taking a feminist approach highlights how structural and spatial inequalities mediate the production (and non-production) of information in a given domain. Second, we deploy this analytic to examine geographies of missing data about feminicide at three levels. To begin, we show that official case documentation about feminicide is spatially uneven, marked by underreported and unreported information. Next, we discuss how this spatial unevenness is produced by socio-political processes that shape information production. In particular, the spatial reach of state and media institutions—based on their presence, infrastructures, resources, and political support—tends to shape those spaces and bodies that are visible, mappable, or recordable—and those that are not. Finally, we explore how activists challenge these institutional and narrative forms of spatialized neglect through creative informatic and organizational strategies, thus making visible and countering the “[l]ayers of taken-for-granted gendered and racialized geographies [that] make unremarkable the abandonment of certain groups of women” (Pratt, 2005: 1067). We show that activist strategies are both place-specific and transcalar. Data activists are often simultaneously operating at and across hyperlocal, municipal, national and international scales. Activists make visible the locations and context of violence; they create coalitions to monitor feminicide in los territorios; they use their databases as sacred sites of memorialization; and they build transnational networks of solidarity and knowledge-sharing. In this sense, we argue that counterdata production about feminicide constitutes multiple spaces of resistance, which reframe and resignify everyday forms of gender-related violence, including by navigating and illuminating geographies of missing data.

Feminicide, data and space

Our analysis places feminist and information geographies scholarship in conversation with literatures on data activism and data feminism. Feminist geographers have highlighted the ways in which space is both constituted by and constitutes various forms of violence, thus inflecting socio-political action around it (Damián Bernal and Ibarra García, 2020; Fuentes, 2020). Data feminism is a framework for making visible and challenging such power asymmetries in the context of data production, analysis and circulation (D’Ignazio and Klein, 2020). Inspired by intersectional feminist thought, it seeks to both understand and address data harms, undertaking data science in the service of social change. In this sense, data feminism shares with feminist geography an attention and commitment to “transformative, disruptive activist theory and practice” (Faria et al., 2020).

Our interest in the relationships between space and the production of counterdata on feminicide builds on an extensive body of work demonstrating the role that territories and socio-economic landscapes play in everyday gender-related violence (Fuentes, 2023; Monárrez Fragoso et al., 2010; Segato, 2004). Space, indeed, can have multiple meanings. With Massey (2005), we understand space as a product of social relations and practices, mutually “constituted through interactions, from the immensity of the global to the intimately tiny” (p. 9). The spatial dynamics of feminicide can be read through the Latin American Indigenous and decolonial feminist concept of cuerpo-territorio [body-territory] that speaks to “the inseparable ontological relationship between body and territory” (Zaragocin and Caretta, 2021: 1504, see also Cruz-Hernández, 2016). As Segato (2004) expresses, “feminized body” also means “territory”: feminicide is a mark on the feminized cuerpo-territorio and a mark of territorial control (p. 12). But cuerpo-territorio is “not only a site of violence and dispossession but also a functional and symbolic space where life, care, and meaning are (re)produced” (Glockner et al., 2024: 7). In this sense, activist data practices can be ways to inhabit, make visible, and care for cuerpo-territorios affected by feminicide.

Spatial aspects of feminicide have been analyzed in relation to the rollout of neoliberal capitalism in Latin America (Berlanga Gayón, 2015) and narcopolitics and processes of gentrification in Mexico (Wright, 2013). Scholars have shown, for example, how spatialized economic violence played a significant role in the murders and disappearances of young women in Ciudad Juárez. Poor women who lived in informal areas of the city and relied on public transport to reach their work sites, often in the maquilas that proliferated following NAFTA, were disproportionately targeted (Monárrez Fragoso et al., 2010). In its spatial dimensions, feminicide has also been shown to intersect with forced migration, state violence and market violence in Central America (Varela Huerta, 2017), conflict in Honduras (Jokela-Pansini, 2020), Indigenous land theft in North America (Simpson, 2017), extractivism and the silence around Indigenous feminicides in Argentina (García Gualda, 2020), and the emergence of financialized modes of extractivism and gendered relations of debt (Cavallero and Gago, 2021). The stay-at-home and lockdown orders of the coronavirus disease 2019 pandemic also had a spatial and gendered dimension (see for example, Rose-Redwood et al., 2020). They burdened women with extra care responsibilities, trapped them in homes with abusive partners, and closed or severely curtailed support services, leading to a rise in gender violence across contexts. This body of work underlines, as Nolin and Fraser (2015) argue, that spatial analysis expands our understanding of feminicide, providing “both tools of solidarity and empowerment to women in situations of fear, as well as the reinforcement of evidentiary support for the implementation and enforcement of law” (p. 31).

In contrast to some of this work, in this article, we are not concerned with understanding the political-economic geographies that produce or reinforce gender-related violence and feminicide nor in comparing violence across locations. We believe the literature cited above already offers many insights in this regard. Rather, our interest is in understanding how space structures or mediates the production and availability of information about feminicide, in relation to information produced (or not) by the state, the media, and data activists. This speaks to an emerging research area called “information geographies,” which explores the spatial distribution of information access, creation, curation, and representation, as well as the power dynamics that shape those spatial patterns (Graham et al., 2015). In this sense, violence against women can be understood as a spatial phenomenon produced not only in specific acts of violence, but also in the (lack of) actions of prevention, attention, access to justice and responses to eradicate it. These actions and inactions, including (spatialized) data and information practices implicate activists, states, the media, and others.

Within planning and geography, there is a long history of work that focuses on knowledge production by the state, and specifically on data production about citizens and populations. In Seeing Like a State, Scott (1998) classically claimed that the state uses data and mapping tools to render populations and territories legible and governable. However, the state does not see nor value all spaces and populations equally. There are important phenomena that states ignore or fail to care for, particularly in relation to minoritized groups. Building on the work of Ọnụọha (2016/2023), D’Ignazio and Klein (2020) use the concept of missing data to illuminate these dynamics. They define them as data that are neglected by institutions, despite political demands that such data should be collected and made available. Examples include eviction tracking, police killings of citizens, and maternal mortality databases, among many others. Further work has shown that missing data may include data that are entirely absent but also data that are sparse, overlooked, poorly collected and maintained, purposefully removed, difficult to access, infrequently updated, contested, and/or underreported (Jungs de Almeida et al., 2023). In the case of feminicide, we propose that missing data have important spatial dynamics—where feminicide goes neglected and uncounted is a key dimension for understanding whose lives and deaths go neglected and uncounted.

Recent decades have seen increasing scholarly attention to the harms of state and corporate data production, as well as the use of data and artificial intelligence in institutional decision-making. There is also a rising interest in alternative epistemologies of data science. Scholars have used the terms “citizen data” and “data activism” to conceptualize the use of data and software by citizens, residents, and social movements to support collective action and exercise political agency (Fotopoulou, 2019; Milan, 2017). The data feminism framework put forth by D’Ignazio and Klein (2020) articulates seven feminist principles for working with data, two of which are particularly relevant for understanding and engaging with data activism: examine power—understanding how inequality shapes the production and quality of data and information about a given topic—and challenge power—using data to mitigate, disrupt and dismantle systemic oppression. These principles are intrinsically linked with the analysis of missing data. Because institutions have often failed to adequately document cases of feminicide, feminist and women data activists have stepped forward to undertake the difficult and painful labor of assembling case records and memorializing lost women (Wright, 2009).

Most activists do this work by manually and systematically reviewing news articles, triangulating those with official data (if any exists in their context), and recording case details in spreadsheets and databases. In 2014, Dalton and Thatcher (2014) drew from work on counter-mapping—the use of maps to produce alternative narratives and representations of spaces, places and people—to theorize “counterdata actions.” In our own prior work, we built on this idea to show that feminicide data activism constitutes counterdata production: data that are produced by civil society groups or individuals not only to challenge missing data but also to challenge state bias and inaction, galvanize media and public attention, reframe political debates, work towards policy change, and help heal wounded communities (D’Ignazio et al., 2022). This is to say that activist counterdata production about feminicide is not only about “filling in the gaps” for state management of populations nor only about denouncing the injustices of powerful institutions.

Activists use systematic information for memory, healing, and social transformation as well as to create “possible alternative data frameworks and epistemologies that are respectful of populations, cultural diversity, and environments” (Ricaurte, 2019: 352). This resonates with Sletto et al.’s (2023) concept of “other mappings”, which they propose to “[encapsulate] the diversity of processes, media, and representational forms of such subaltern representations and the ways in which they point to situated relationships between land and body” (p. 149). In the case of feminicide data, as we will discuss, activists use spreadsheets and databases, and in many cases also maps, to illuminate situated relationships within cuerpo-territorios, not only between land and bodies but also between data and bodies.

Related to the spatial dynamics of information production, Ricaurte (2019) uses the terms “data extraction” and “data colonization” to highlight forms of exclusion and erasure emerging from data collection processes and data-centric epistemologies when they are analyzed as sites of colonial power. Ricaurte (2019) explains that the current extractive power structure of data entails the “violent imposition of ways of being, thinking, and feeling that leads to the expulsion of human beings from the social order, denies the existence of alternative worlds and epistemologies, and threatens life on Earth” (p. 351). There is a propensity for this uneven power structure to be embedded within conventional data collection practices. As we will show, data activism about feminicide is not working within dominant quantitative approaches to counting fatalities. Activists use spatial strategies to challenge “disembodied epistemologies” and establish proximity between “those killed and those watching,” a feminist geopolitical concept that Hyndman (2007) has called “moral proximity.” Their data practices manifest Williamson’s (2021) aspiration “to think numbers, data, and data collection differently than is typical of the vast majority of work on serial murder” (p. 27).

Methods

Data Against Feminicide is a collaboration between Feminicidio Uruguay, an activist initiative, the Latin American Initiative for Open Data, a nonprofit organization, and the Data + Feminism Lab, a research laboratory. The project supports feminicide data activists through knowledge-sharing, participatory technology development and community building. ¹ The analysis we present here draws specifically on one strand of the broader project, which is a qualitative inquiry into activists’ data practices. To this end, we have interviewed 33 activists across 15 countries who use data in their work on fatal gender-related violence, mainly in Latin America (see Supplementary Appendix 1).

The interviews were selected based on a purposive sampling strategy designed to understand both activist data production efforts that explicitly focus on feminicide and those that attend to gender-related killings or disappearances in their intersections with racialized, transphobic, settler colonial, and ethnic forms of oppression. Our geographic focus reflects the interests and lived experiences of the research team—which includes one co-author who is both part of the project’s leadership and herself an activist. In particular, our research team wanted to center the strength of the Latin American feminist movements’ global agenda setting around the problem of feminicide, and the genealogies of activism noted earlier. We also included six interviews with data activist groups in Northern America to counteract colonial and imperial perceptions from the Global North that fatal gender-related violence is a problem that happens “elsewhere”. These framings are particularly incorrect and painful in the ways they elide the crises of MMIWG2S and police killings of Black women, as well as diminish long-standing activist efforts in the US and Canada to make visible fatal gender-related violence against women (Collard, 2015; Williamson, 2021).

Our approach to the interviews built on data feminism’s commitment to challenging expert power hierarchies in knowledge production and to “embracing pluralism” by centering the lived experiences, practices, and goals of diverse activists and organizations engaged in counterdata production on feminicide and fatal gender-related violence (D’Ignazio and Klein, 2020). Interviewees had various backgrounds (e.g., social work, journalism, nursing, academia, law) and typically led and/or were directly engaged in monitoring efforts. The interviews were conducted via video conferencing in Spanish, Portuguese, and English depending on the interviewees’ language. They lasted roughly two hours and followed a semi-structured format that covered interviewee/organization background, data collection process, categorization, and reflection on activism. While we did not set out to examine the spatial dimensions of feminicide data activism, we noticed after initial interviews that spatial themes consistently emerged in activists’ accounts of their monitoring practices. In subsequent interviews, we added questions that surfaced spatial dimensions such as which regions or places were more difficult to find information about, the geographic characteristics of those places, and mapping approaches activists used to visualize spatial data.

As part of our analysis, we want to highlight, first, that our reading of the interviews centered on understanding the spatial aspects of data production—that is to say, the geographies of information production about violence rather than geographies of violence themselves, though we recognize the two are inextricably linked. Second, our approach does not aim to compare projects or places but rather seeks to illuminate “commonalities across differences” (Mohanty, 2003). In our case, this consists of highlighting common spatial practices and challenges around feminicide data that surface for activists despite the significant variation we know exists among their contexts. The spatial themes we discuss were apparent across transnational feminicide data production, even as individual data activism efforts are situated, forge place-specific categories for feminicide, and are deeply connected to the history, culture and legal frameworks of their places.

Geographies of missing data

In this section, we explore how information and knowledge about feminicides is produced and circulated, and the ways in which these processes create geographies of missing data. Data production is a situated and relational process, taking place within information ecosystems that shape the generation and circulation of information and knowledge. Information flows across scales and sites as it is produced, curated, transformed and used by actors within such ecosystems (Susman-Peña, 2015). In our research, when activists explained how they learned about new cases and the challenges they encountered in their monitoring efforts, their accounts often brought to light how available data about feminicide are spatially uneven.

One important issue concerns reporting. It is well-established that gender and sexual violence is underreported due to stigma, victim-blaming, lack of trust in institutions, and hostile reporting climates (Sovereign Bodies, 2020; UNODC, 2019: 40–41). Amongst those cases that are reported, activists’ monitoring experiences revealed that some cases or stories tended to be more legible, visible, or “mappable” than others, especially when these “others” intersected with forms of socio-spatial segregation and inequality. Geographies of missing data thus are produced through various intersecting socio-political processes—involving the state and the media, but also organized crime or communities/families—that shape information production and circulation. Such patterns sometimes respond to processes where “people are rendered as bare life and legally abandoned” (Pratt, 2005: 1054).

When searching for information about feminicide, the majority of activists and journalists look first to the state. In some cases, they encounter the sheer absence of official data. This is what the Red Feminista Antimilitarista found when they requested statistics about women’s murders in Medellín (Colombia), “We realized that our institutions didn’t have information systems.” More often, however, activists find official data to be replete with problems like undercounting and misclassification. One reason for this has to do with legal frameworks and classification practices, as well as issues with judicial processes. In Canada, a leader from the Canadian Femicide Observatory for Justice and Accountability (CFOJA) stated, “nothing is officially seen as a femicide because we don’t have any legislation or any official recognition of femicide.” In countries where legal frameworks do exist, they are often narrowly framed around intimate feminicide, where the perpetrator has a pre-existing relationship with the killed woman, and often exclude trans women (D’Ignazio, 2024). Moreover, judicial processes can be flawed and lengthy. In Brazil, Néias - Observatório de Feminicídios de Londrina focuses on monitoring and reporting on court procedures and rulings regarding potential cases of feminicide, observing that some cases take years in court before being officially treated as feminicides, while others may be misclassified due to low-quality evidence from police investigations. A related issue is state misclassification of gender, ethnic and racial minorities. Activists described how trans women are nearly always misgendered by law enforcement in death. Groups that monitor Indigenous feminicide also repeatedly highlighted the misclassification of both race and cause of death for Indigenous people on death certificates.

Problems with undercounting and misclassification in official data also often reflect uneven state capacity to investigate and record cases. The Alianza Feminista para el Mapeo de los Femi(ni)cidios en Ecuador, for example, noted a critical absence of the state in rural areas, leading activists to speculate that cases of feminicide in those areas are much higher than official counts. In Northeast Brazil, journalists from Uma por Uma observed that high police turnover and lack of expertise in smaller municipalities tend to hinder the investigation of feminicide cases. In Costa Rica, feminist researcher and activist Mariana Mora of RECORDAR-LAS noted that the state simply “doesn’t exist” in remote areas that can only be reached by boat. This illustrates how problems with state capacity relate to infrastructural problems, including inadequate, precarious or inexistent physical (roads, services, state offices) and digital infrastructures (internet and telecom services) that support information production and circulation across geographic spaces.

To some extent, systemic gaps in information from rural and remote areas seem to follow an established dynamic in information geographies where comparatively more data about a phenomenon are produced in urban centers with robust infrastructure than in rural, remote and marginalized places in a given country (Graham et al., 2015; Walker and Frimpong Boamah, 2017). But, as we discuss further, there are also patterns of underreporting of feminicide in marginalized areas in urban centers, in places where ethnic, racial and gender minorities live or work, in territories controlled by organized crime, and in territories disrupted by mass migration, environmental disaster or conflict. Thus, geographies of missing data are produced across different landscapes of the cuerpo-territorio, and are shaped by the state’s reach, capacity, and interest in (or neglect of) a given area or territory.

Because states fail to produce and publish comprehensive official information about feminicide, activists often turn to the media. The vast majority of activists we interviewed rely on news articles as primary sources for identifying cases of feminicide, verifying case details, and recording cases in spreadsheets or databases. Triangulating official sources of information with media reports has enabled activists to demonstrate the systematic underreporting of feminicide by the state—notably in Puerto Rico, Guatemala, Mexico and Brazil. Yet, activists have a fraught relationship with news media coverage of feminicide, which they consistently characterized negatively as “sensational”, “victim-blaming”, “stigmatizing”, “transphobic”, “misogynist”, “racist”, and “trauma porn.” Many described how the media amplifies state biases through relying on law enforcement as an authoritative source of information. One example is the state’s misgendering of trans women in death which leads to what one activist framed as a “double murder”: the person’s lived identity is murdered again in the media coverage of the event. Activists also described how news articles are not a comprehensive antidote to the state’s missing data. Many noted that media outlets tend to regard as less newsworthy those cases of feminicide where the victims belong to marginalized gender, racial and ethnic groups, frequently neglecting to report on them. For example, a member of the Coordinadora Nacional de Mujeres Indígenas (CONAMI, Mexico) described how the feminicides of Indigenous women are “the most invisible of the invisible.”

Media bias is often compounded by the uneven spatial presence of media organizations themselves. Similar to the state, the offices and staff of larger media outlets tend to be concentrated in urban centers. According to the Red Feminista Antimilitarista, “the feminicides that take place in the center of the country are well documented. But those that happen on the periphery are not.” Moreover, industry pressures have led to the shuttering of local news services of all kinds, so media’s geographic reach is contracting in many countries. For example, the executive director of Jane Doe, in Massachusetts (United States), told us it had been “harder in the past few years to get information from rural communities because the local papers and local media have shut down in so many places.” Yet, important information about cases can go unrecorded in central urban areas as well. For example, Justice for Native People (United States and Canada) observed that it can be equally hard or harder to locate cases of fatal violence against Indigenous people in urban areas because law enforcement and the media often misclassify Indigenous people or entirely fail to record the race/ethnicity of the victim.

As the examples above indicate, forms of institutional absence, bias and omission as well as infrastructural limitations lead to systematic geographic patterns of neglect in information production. In other words, they produce geographies of missing data. These include missing data about feminicide in rural and remote areas, as well as about ethnic, racial, or gender minorities. This may encompass large territories with high populations of racial or ethnic minorities, such as tribal reservations or traditional Indigenous lands, or smaller scale sites such as urban areas with informal settlements, racially and economically segregated neighborhoods, factory work sites of migrant women, or the urban geographies of trans sex workers.

In some instances, families and communities can be important sources of information about individual cases of feminicide. However, activists are careful in how they engage with them. The majority do not contact families directly but might work with and share information when families contact them first. The primary way in which families contribute information about feminicide is through their decision to report or not report a case to the state or in the media. Activists in Latin American countries, for example, noted that it is common for families to seek justice through reporting and making their case widely known in the press. However, there are several reasons why families may feel discouraged from reporting. An activist from Black Femicide US noted that the rural communities that she works with dislike engaging with institutions outside the community, “They do not like to involve outsiders in what they consider family business, whether you’re related to them or not.” Families from minoritized communities—such as racial minorities, trans women, undocumented people, and sex workers—may face violence and retribution for engaging with law enforcement. Organizers of the #SayHerName campaign wrote about the case of Kayla Moore, a transgender woman in California who was undergoing a mental health crisis (Crenshaw and Ritchie, 2015). Her roommate called the police who, instead of providing support, attempted to arrest her and ended up suffocating her to death in her own bedroom.

If the feminicide occurred in the context of organized crime, families may face further violence or intimidation from the perpetrator or his organization for reporting to officials. Activists mentioned organized crime—including gangs, armed men, paramilitary activity, and hitmen—both as a perpetrator of feminicide and as a disruptive factor on the availability and accessibility of feminicide information. While the origins and nature of organized crime vary across contexts, their activities are often spatialized through territorial control (Fuerte Celis and Pérez Lujan, 2017). Where organized crime is widespread, it can suppress information flows across the whole ecosystem, hindering media coverage, investigation by state actors or family reporting. As María Salguero, the creator of Yo te Nombro: Mapa de Feminicidios en México, said of Mexican regions controlled by criminal groups, “Sometimes journalists, out of fear, don’t want to narrate in detail, nor even mention the method of death. Sometimes all they say is ‘she was executed’.” Likewise, activists from Colombia Diversa described the risks for families, “Reporting cases of violence puts people in the crosshairs of armed actors.” One of the effects of such widespread intimidation and territorial control is that feminicides that occur in places dominated by organized crime also tend to be systematically underreported and/or misclassified.

Other factors that affect the production and circulation of feminicide information may include situations of civil war, economic or political collapse, and natural or climate-related disasters. While they did not emerge as frequently in our interviews, such crises are well-documented as disruptive forces to stability and to physical and digital infrastructures, and have been known to increase feminicide (Thurston et al., 2021). They also often lead to mass migrations, with women refugees being more vulnerable to violence and trafficking. In the case of Venezuela, an activist from Utopix associated the brutal assassination of two prominent women who led efforts to distribute food to people in need with the country’s unstable political and economic context and with organized crime groups seeking to control territories that the state had abandoned. Such social and political turmoil and disruption of reporting mechanisms, infrastructures, and services may also produce geographies of missing data through systematic underreporting of feminicide information.

Spatializing counterdata production

Activists navigate and creatively confront geographies of missing data on feminicide through various evolving counterdata strategies. In this section, we explore the spatialities of such strategies. We draw attention to four interrelated modes of spatialization: reach, emplace, reclaim, and cruzar (crossing).

Reach

By reach we refer to counterdata strategies that seek to amplify activists’ ability to identify cases and access information from different parts of the territories they cover. These strategies emerge to make it possible to understand and monitor feminicides in rural and remote areas as well as marginalized urban areas. For some groups, “reaching” a remote territory begins at the organizational level. Most of the activists we interviewed are based in urban hubs, which shapes their ability to access information about different places. Some groups deal with this by adopting a decentralized model of organization, establishing different satellites at regional, province or neighborhood levels. For example, in Argentina, Mumalá shifted from a centralized to a decentralized model in 2018, when they established working groups in each of the 20 provinces where they monitor feminicide cases. Another strategy for amplifying reach consists in establishing geographically dispersed structures and coalitions. For example, the CFOJA has an advisory group of 35+ women based across Canada. Similarly, the journalist-led project Uma por Uma in Northeast Brazil often relied on their network of journalists spread across the state of Pernambuco to learn about and gather information on new cases in rural or remote locations.

Several groups also crowdsource information through social media, messaging applications and organizational announcements, building on connections with other feminist, women’s and community organizations. Given the aforementioned forms of state and media bias, these strategies are particularly important for groups that monitor intersectional forms of feminicide such as transfeminicide and Indigenous feminicide. Indigenous networks, for example, have been important for reporting cases that happen in remote communities that are not covered by the state, media or social media. An organizer from Red Feminista Antimilitarista mentioned that “we are aware of Indigenous feminicide cases only through the organizations’ official announcements”. Along with crowdsourcing, many activists consult hyperlocal media sources—from small newspapers to local TV stations, blogs and radio stations—to help with identifying cases in rural or inland areas that national or mainstream media rarely cover. A leader from the Alianza Feminista para el Mapeo de los Femi(ni)cidios en Ecuador observed that the most important sources for learning about rural and remote cases are “the Facebook pages from local media, they are tiny. There are many here in Ecuador.”

Such strategies to amplify reach, however, require continuous effort and remain necessarily partial. Activists articulated to our team how, even as they try to find inventive ways of learning about the stories, they themselves are inevitably also producing missing data. They know that some geographies remain less documented than others and that their databases are not comprehensive archives of feminicide cases. As a staff member at the African American Policy Forum said, “we know that there are so many that we are missing.”

Emplace

The emplace metaphor relates to how activists collect, produce and visualize geographic data in order to situate feminicides in place and in context. The majority of activists collect some spatial data about every case. This ranges from general geographic information about the location of the murder, such as town, city, and municipality, to more precise geolocation such as x and y coordinates, addresses, or zip codes, and details about whether the violence occurred in a home, a workplace, or public space. Locating cases in space helps to illuminate the structural nature of this violence. But emplacing is also about using spatial data to understand the contexts in which specific cases occur and to challenge narratives about murdered women. This contextual analysis helps activists infer whether a case is a feminicide and what type of feminicide it is. Moreover, it helps them to challenge conventional socio-spatial representations, like the idea that homes are “safe spaces” or the stereotype that feminicide occurs more frequently in poorer or marginal areas.

Those contexts constitute what Carcedo (2011) has called “scenarios of femicide” or “the socio-economic, political and cultural contexts in which particularly uneven power relations between men and women are generated or deepened and where this power imbalance leads to issues of control, violence against women, and femicide” (p. 15). The modus operandi is thus ascribed not (only) to the killer or killers but to the societal context in which the murder takes place. Many activists work with typologies of violence that make these relations explicit. They may create their own categories or draw from existing classifications, like the Latin American Model Protocol for the investigation of gender-related killings of women (femicide/feminicide), which enumerates 13 types of femicide, including intimate femicide, racist femicide, femicide because of trafficking, and transphobic femicide, among others (United Nations High Commissioner for Human Rights, 2014). In these scenarios of femicide, the locations of bodies and places of death offer clues about the context of cases and thus which types of feminicide they constitute. The creator of the Yo te nombro map, for example, said bodies are often found in hidden places such as “extermination camps where they burn bones and people until they disappear”, a pattern she links to organized crime “because people from the Jalisco cartel say: ‘No body, no crime.’” Likewise, a staff member from CFOJA observed that “there is heightened gang violence in Western Canada, especially in BC,” because there are a lot of murders in public scenarios associated with gang violence due to affiliations or because women are treated as a means to exert domination over a territory. Activists from the Red Feminista Antimilitarista (2015) use the category “neoliberal feminicidal violence” to explain how the territorialization of violence against women, as they see in Medellín, is fueled by neoliberal logics and the attendant convergence of narcoparamilitary, military, and armed private men. Furthermore, activists’ aggregation of data allows them to surface and illuminate spatial patterns for specific cases. For example, activists from Letra Ese discovered a high incidence and spatial pattern of bodies of trans women and lesbians found in public open spaces such as rivers, water canals, roads, or even cemeteries, or, in the case of sex workers, near their places of work.

To understand the geographies of violence in their territories, activists engage in spatialized data practices that situate relationships between land and body. At least a third of the groups we interviewed geolocate and/or visualize cases in the form of a map. In this sense, activists’ (spatial) data practices are “a feminist means to diagnose territorial conflicts and to initiate healing of bodies and territory” (Zaragocin and Caretta, 2021). For example, the creator of Ellas tienen nombre told us she initially decided to make a map to see “where she is standing” in relation to the gender-related violence that occurs in her territory, though her motives for continuing became more complex over time. Activist María Salguero conducts a tailored analysis of geolocated feminicide data for each region in Mexico. She succinctly states the relation between geographies and violence: “the conditions of each territory shape violence against women.” Other groups visualize this relation through their categories. For example, CONAMI works with the category of ecofeminicidio to describe the assassination of women land defenders. And indeed, they see all violence against Indigenous women in Mexico as linked to the defense of Indigenous lands and cosmovisions.

Activists recognize the inherent tensions in emplacing violence, particularly through maps (intimately linked to the violent history of colonization), since the visibility granted to specific locations through data may render the bodies who inhabit those spaces susceptible to stigma and further oppression, such as when discourses against feminicide and gender-related violence become a pretext for military action or policing of specific populations (Fuentes, 2023: 72). Accordingly, activists carefully weigh up these tensions when visualizing geographic data. For example, the group Cuántas Más initially produced a map to contextualize feminicide cases in Bolivia, but they decided to take it offline when they realized that it risked re-stigmatizing vulnerable areas and thus re-inscribing the same forces of domination that feminist activism seeks to challenge. In the case of Feminicidio Uruguay, the mapper discussed how concerns about vigilantism influenced her decision on which data to collect and publish about perpetrators. While activists see the map as a powerful evidence-based tool to argue with the state and to link bodies and territories, they also described the risks and limitations of mapping for visibility.

Reclaim

Reclaiming refers to spatial strategies that aim to memorialize and contextualize feminicide as a structural phenomenon. First, feminicide data activists undertake public interventions in which they creatively transform their data into striking spatial and physical forms. In some instances, these are large-scale installations. In 2021, the group Mujeres de Negro Rosario planted 250 chairs outside the provincial court building in Rosario, Argentina, a setting specifically chosen to interrogate power and demand justice. Each chair was labeled with a woman’s name and information about her feminicide. As one organizer stated, “Seeing people’s faces … I mean, one says 250, and it’s a number, but when you see that represented with the space that woman occupied, this has another impact on people.” In other instances, interventions are smaller in scale but targeted, such as when activists from Néias in Brazil camped outside a local court “to give a message to the judiciary that society was closely watching the trial” of a feminicide case. These takeovers of public space often happen in parallel with important dates, such as International Women’s Day or the International Day for the Elimination of Violence against Women. This spatio-temporal conjunction represents a way to also reclaim media attention, since media seek to report on newsworthy events related to the date.

Other forms of claiming space are quieter and more reflective. For example, the group Movimiento Manuela Ramos hosts a monthly vigil where they produce a large-scale (approximately 2 m × 2 m) calendar with all of the names of feminicide victims in Peru from that month placed on the day they were killed. As part of the remembrance ceremony, activists place their handprints on top of the handprint on the calendar next to each woman’s name. One of the co-leaders described the meaning for their group,

Each one of us would reach out and put our hand on that little hand as if we were trying to make a connection. More than a connection, it’s like—each one of those victims is an inspiration of life for us.

In these cases, spatializing their data allows activists to craft rituals of remembrance, claiming space and time for commemorating collective loss. These acts of reclaiming represent what Delia Duong Ba Wendel (2025) has called “an ethics of non-erasure”, the crafting of representations of violence into spatial form to tell otherwise ignored, hidden, and neglected truths.

Finally, a more subtle spatialization that emerged from our interviews is that most feminicide data activists treat their databases themselves as sacred sites of memorialization. A member of Sovereign Bodies Institute described how families request that their loved one be included in the database,

Entering [a person] in the database doesn’t necessarily bring their case to justice and it isn’t justice for them or their family on its own, but it does feel like we’re giving them a little bit of peace knowing that they’re being counted and they’re being honored and they’re being prayed for.

While the received wisdom in data science is that a database is a space of “raw” or latent information, awaiting activation by analysis or visualization, feminicide data activists reclaim database technology to do exactly the opposite. The database in this case becomes a space for collective witnessing, a refusal to forget an individual and an insistence that their story is linked to larger patterns of structural violence.

As the founder of Women Count USA described,

I felt that if I could bring all of this data into one place […] it could tell a story about what was really happening to women and where it was happening … it would [also] memorialize these women […] It would show that they were more than just statistics on a page.

For her, the database is a place for both honoring individual lives and for asserting the vast and structural scope of the crisis. The place-making labor of feminicide database production stands in sharp contrast to how cases are treated by the media and the state—as individual cases, separated and distinct from each other. Thus, acts of informatic aggregation can also be seen as creating spaces for collective witnessing and memorializing—a key strategy to recuperate and connect those sites where feminicides are systematically underreported or misclassified. They produce spaces that activists see as deeply connected to the embodied lives and people who are partially represented by the data.

Cruzar/crossing

Our last theme, cruzar or crossing, highlights how activists’ counterdata strategies often involve building connections across territories, practices, and experiences. Our use of cruzar/crossing borrows from the idea of cruzar datos/crossing data put forth in a call for Latin American and North-based activists and scholars to collectively reflect on “engaging with data across the borderlands of disciplines, practices, migrations, and languages” (Alvarado Garcia et al., 2022). We find that this captures the spirit of what some activists seek to do with the data they produce on feminicide: to cross borders, to be at once place-based and transcalar, to forge “multidirectional crossings and movements of feminist ideas, practices, and embodiments” (Zaragocin, 2021: 237).

In this context, crossing strategies have various aims. One is to connect data across geographies, illuminating patterns of violence at different scales. Sovereign Bodies Institute, for example, began to document Indigenous women from North America who are missing and murdered in South America and vice-versa, thus attending to patterns of intersectional violence that traverse borders. In a different vein, MundoSur convened activists across the continent in a network (Red Latinoamericana contra la Violencia de Género) as well as a participatory process to systematize their data variables and standardize categorization schema. From that work, they produced a map of feminicide data from different countries in Latin America as a way to influence policy responses regionally. In other cases, such as the Alianza in Ecuador as well as Indigenous activists in North America, a central goal of data collection efforts is to link gender-related violence with “environmental and territorial violence” related to extractive practices such as mining and oil exploration.

The most prevalent way in which activists engage in crossing, however, is through efforts to relate local struggles to regional and global ones, particularly through creating or participating in transcalar networks of knowledge-sharing and solidarity. Both Letra Ese and Colombia Diversa, for example, coordinate the network Red Regional de Información sobre Violencias LGBT across 11 countries. A central goal for them is to bolster organizational capacity to produce data and to foster ties among organizations so they can strengthen one another. Many of the activists we interviewed participate in some form of network or collective space as part of their monitoring practices. These can be more or less formalized, as well as more or less transcalar. An important dimension of these networks is that they create connections across experiences and contexts. They facilitate the exchange of data practices and resources, and they also open spaces of mutual support in the labor—especially the emotional labor—of monitoring feminicides and gender-related violence.

Conclusion

In this article, we have put forward the notion of “geographies of missing data” as an analytic lens grounded on feminist principles to illuminate spatial aspects of data production, focusing on the example of feminicide data. We have demonstrated the ways in which the production and circulation of information about cases of feminicide is spatially uneven, creating places characterized by systemic underproduction or non-production of information. While such sites may vary from place to place, there were some striking transnational patterns of missing data reported by activists, namely the systematic underreporting of feminicides in rural and remote areas; marginalized areas; places inhabited by ethnic, racial and gender minorities, including those who work in stigmatized occupations; territories controlled by organized crime; and territories disrupted by mass migration, environmental disaster or conflict. Space, therefore, is central to the politics of data production about gender-related violence.

The geographies of missing data we describe are primarily produced at the country and subnational scales. They can be traced to abandonment, inaction or bias by the state and the media, the presence or absence of legal, physical, and digital infrastructures, and disruptive factors such as organized crime, war, and climate-related disasters. Where feminicide goes under- and unreported tends to correspond to known dynamics in information geographies, as comparatively more data about the phenomenon is produced in urban centers with robust infrastructure and comparatively less in rural, remote and marginalized places (Graham et al., 2015; Walker and Frimpong Boamah, 2017). Yet, our work also demonstrates that such designations can be blurry and are far from static. When we consider the work of feminicide data activists to recuperate geographies of missing data and work across scales (from local to transnational), it becomes necessary to rethink and complicate spatial binaries, such as rural/urban or core/periphery. This is in line with the principle of data feminism to rethink binaries and hierarchies.

We have also shown how activists counter unevenness in the production and circulation of feminicide information through various spatial strategies, illuminating how counterdata production can be both place-specific and transcalar. One of these strategies is to amplify reach. Activists document feminicide cases in rural and remote territories through spatialized monitoring or governance structures. Yet, activists’ strategies are not only about making geographies of missing data visible, nor only about producing more accurate, spatially even statistics. They also include emplacing, or the careful collection and analysis of geographic information at the scale of the individual case of feminicide. This permits activists to more deeply examine power—to understand context, establish links between bodies and territories, and to infer whether a particular event constitutes a feminicide. Here, activists assert spatial data as key linkages between cuerpo and territorio. Thus, while Ricaurte (2019) emphasizes how data can operate as a force of centralization, reduction and abstraction that enacts dominant power dynamics, systematizing case information here functions as a connective force to ground the case in its spatial context, to place it in the territory, to render it visible, and to name it as part of the larger structural phenomenon of feminicide and gender-related violence.

This naming is intimately linked with reclaiming—using spatial strategies to link individual cases together and to demand collective memory justice. This “ethics of non-erasure” is visible through large- and small-scale interventions in public space using feminicide data and stories to assert the presence of a collective political demand. The ethics of non-erasure is also employed within the databases themselves. Databases and spreadsheets become spaces of memory and witnessing, places where feminist activists—through tending to and caring for the specifics of each case—enact and practice “moral proximity” (Hyndman, 2007) between those who experience injustice and those who witness it. Thus, activist databases themselves become spaces of resistance and resignification of gender-related violence.

In these ways, activist strategies with feminicide data are affective, narrative, and dialogical. They resonate with the data feminism principle to elevate emotion and embodiment—using data to connect to bodies, to beings and to lived experiences. While activist databases are aggregations, they are not about rendering populations and territories legible for management by the state (as Scott articulates in Seeing like a State). Rather, the point of aggregating cases is narrative and affective—placing cases together in the rows of the spreadsheet or database links each individual case to the others and frames them, names them, collectively, as part of the structural and insurgent concept of feminicide. Like a member of the Grupo Guatemalteco de Mujeres stated “It’s not about having data just to have it. It’s about putting it into action in such a way that it can generate a response.”

Finally, while activist databases are place-based, activists reach beyond their own geographies—what we called cruzar/crossing—to build transnational networks of solidarity with other data activists. These transcalar crossings constitute a key method for activists to examine power—comparing analyses of feminicide across contexts and scales, developing new categories and concepts—as well as to challenge power—by creating spaces for mutual support that enable activists to sustain their labor in the face of its many political, financial and emotional burdens.

While several works have described the interrelations of feminicide and space (including numerous articles within this journal), this study reveals how space is also intimately connected to data about feminicide—its production, analysis and circulation. With this article, we extend an invitation to critical data studies to more deeply engage with the spatial politics of data and information. As we have shown, these are intimately bound together in both the geographies of missing data and activist strategies to challenge them. Likewise, we see an opportunity for feminist geography to engage with and enrich the emerging study of information geographies using the principles of data feminism. Bringing a feminist perspective to information geography would help to expand and complicate its consideration of scale and to actively work towards transformation, rather than simply describing the uneven spatial politics of information.

Finally, we would like to return to the mission of the activists themselves. It is important to note that no group we interviewed believed that more data, on their own, would “solve” feminicide. Activists see the production, analysis and circulation of data about feminicide as part of a larger constellation of efforts aimed at transformative change. As a member of the Alianza stated when reflecting on the impact of their work,

We play a minimal role in this process, which is basically a defense, a fight for rights and for a dignified life for women and for the eradication of violence, but this is collective work, networked work, work that has to continue expanding upwards, downwards, towards all the edges that we can give it.

Thus, while this article has focused on the geographies of feminicide data, we acknowledge that activists’ work to challenge feminicide has never been for data, about data, or in service of data. Their work is, and remains, a defense of life, living and vitality for women and gender minorities. We join the vision articulated by a member of CONAMI, “Our aspiration is to eradicate the violence. It is not about prevention, it is not about management. Our dream is to eradicate all of the violences.”

Supplemental Material

sj-pdf-1-epd-10.1177_02637758241275961 - Supplemental material for Geographies of missing data: Spatializing counterdata production against feminicide

Supplemental material, sj-pdf-1-epd-10.1177_02637758241275961 for Geographies of missing data: Spatializing counterdata production against feminicide by Catherine D’Ignazio, Isadora Cruxên, Angeles Martinez Cuba, Helena Suárez Val, Amelia Dogan and Natasha Ansari in Environment and Planning D: Society and Space