Critical Data Studies

Critical Data Studies

Critical Data Studies (CDS) is an interdisciplinary field that seeks to identify how the epistemological and ontological implications of data collection and data-driven processes may (re)constitute both knowledge and subjectivity (Hintz et al., 2019). It is concerned with the significance and power of digital data in contemporary society and how data relates to broader processes of societal transformation (Hepp and Kramp, 2022).

Being critical in this context does not mean simply criticising data or data systems. Rather, it means reflecting carefully on context and validity — interrogating the assumptions embedded in data practices and asking probing questions about specific information, phenomena, and power relationships.

Key Questions in Critical Data Studies

Selwyn (2019) identifies four core questions that organise CDS inquiry:

• How does data alter reality? — How does data shape our knowledge, our social world, our sense of self, and the decisions that govern our lives?
• What are the values of data? — Who benefits from data? What economic, social, or political value is created, and for whom? Whose interests are served?
• What shifts in power are associated with data? — How is data used within systems of power? Who controls data, and who is controlled through it?
• What are the social and political consequences of our data-driven society? — How does datafication differ from other forms of social control, and what new risks or harms does it create?

Data is a Form of Power

The foundational claim of Critical Data Studies is that data is a form of power. Data does not simply represent the world — it actively shapes it. Data systems determine who is visible and who is not, who receives services and who is denied them, who is surveilled and who is trusted. Accepting this claim allows us to analyse data practices alongside existing power structures — including class, race, gender, and colonialism — rather than treating them as neutral technical phenomena.

CDS also investigates the structural barriers surrounding people and their data: What agency do individuals have over their own personal data? Who sets the terms of data collection and use? These questions of agency and consent are central to data ethics and governance debates.

Data Feminism

Data feminism is an intersectional approach to data that examines the relationship between data systems, power, gender, class, race, sexuality, ability, age, religion, and geography. As D'Ignazio and Klein argue, it begins with a belief in gender equality and a commitment to examining the root causes of the inequalities that certain groups face today.

The Seven Principles of Data Feminism provide a framework for both analysing existing data systems and building more equitable ones:

1. Examine power — data feminism begins by analysing how power operates in the world; it asks who collects data, about whom, and why
2. Challenge power — data feminism commits to challenging unequal power structures and working toward justice; it is not enough to describe inequality, we must actively work against it
3. Elevate emotion and embodiment — data feminism values multiple forms of knowledge, including knowledge that comes from people as living, feeling bodies in the world; emotion and experience are valid data sources
4. Rethink binaries and hierarchies — data feminism requires us to challenge the gender binary along with other systems of classification (race, ability, sexuality) that perpetuate oppression through oversimplification
5. Embrace pluralism — data feminism insists that the most complete knowledge comes from synthesising multiple perspectives, prioritising local, Indigenous, and experiential ways of knowing
6. Consider context — data is never neutral or objective; it is the product of unequal social relations, and understanding this context is essential for accurate and ethical analysis
7. Make labour visible — the work of data science involves many hands, including often-invisible care, emotional, and maintenance labour; data feminism makes this visible so it can be recognised and valued

Data Colonialism

Couldry and Mejias (2019) introduce the concept of data colonialism to describe a new form of exploitation that extends colonial logics into the digital age. They define it as combining "the predatory extractive practices of historical colonialism with the abstract quantification methods of computing." Data colonialism operates through three mechanisms:

• Data capture as extraction — data is taken from people's social lives and bodily movements without fair compensation or meaningful consent, mirroring the extraction of resources from colonised territories
• Data-based judgements — algorithmic systems make consequential decisions about individuals and communities based on data, often in ways that reproduce historical inequalities
• Colonisation of the self through data — as more aspects of identity and social life are datafied, people increasingly understand themselves through data categories that were designed by others for others' benefit

Digital colonialism is the related use of digital technology for political, economic, and social control. Big Tech corporations use proprietary software, corporate cloud infrastructure, and centralised services to process users' data for profit — creating a fundamental disconnection between the ownership, control, and location of data versus the people it is about. This dynamic has prompted calls for data sovereignty: the right of countries, communities, and peoples to control and own their own data.

Data Justice

Data justice is a normative framework concerned with ensuring fairness and equity in how data systems affect people's lives. In an increasingly data-driven world, three pillars organise data justice:

• (In)visibility — who is made visible in data (and thus able to access services) and who is rendered invisible (and thus excluded or erased); both visibility and invisibility can be forms of harm
• (Dis)engagement with technology — who has genuine agency and choice over their engagement with data systems, and who is compelled to participate without meaningful alternatives
• Antidiscrimination — preventing data and algorithms from being used to discriminate against individuals or groups on the basis of protected characteristics or social position