Data and the New Forms of Social Inequality
Data is often presented as a tool for progress. It promises better decisions, more efficient systems, and more targeted interventions. In governance, markets, and everyday life, data is positioned as a means to improve outcomes and reduce uncertainty. Yet beneath this promise lies a more complex reality. Data does not only reveal inequality. It can also reshape and reproduce it in new forms.
As societies become increasingly organized through data, inequality is no longer defined solely by income, education, or access to resources. It is also defined by how individuals and communities are represented, processed, and acted upon within data systems.
From Traditional Inequality to Data Mediated Inequality
Traditional forms of inequality are grounded in material conditions. Access to land, capital, education, and employment has long shaped social outcomes.
In a data mediated world, these forms of inequality intersect with new dimensions. Representation in data, access to digital infrastructures, and the ability to interpret and use data become additional factors that influence opportunity.
This does not replace existing inequalities. It overlays them. Those who are already disadvantaged may also be underrepresented or misrepresented in data systems, compounding their exclusion.
Kitchin (2014) highlights how data driven systems can reflect and reinforce existing social structures. Data is not created in a vacuum. It is produced within contexts shaped by inequality.
The Politics of Representation
One of the most significant ways data shapes inequality is through representation.
Data systems rely on categories, classifications, and variables to describe individuals and communities. These representations determine how people are seen within governance and decision making processes.
However, not all experiences can be easily captured. Informal work, precarious living conditions, and complex social arrangements may be poorly represented in data.
Bowker and Star (1999) emphasize that classification systems shape social reality by defining what is included and what is excluded. In a data driven society, representation becomes a condition for recognition.
When individuals or groups are not adequately represented, they may be overlooked in policies and services. Inequality is thus reproduced not only through lack of resources, but through lack of visibility.
Data Bias and the Reproduction of Inequality
Data systems are often assumed to be objective. However, they are shaped by the data they use.
Historical data reflects past decisions, including biases and inequalities. When such data is used to train models or inform policies, these patterns can be reproduced.
Barocas and Selbst (2016) show how data driven systems can produce disparate impacts, even without explicit intent. Bias can emerge from incomplete data, skewed samples, or flawed assumptions.
In areas such as credit scoring, hiring, and public services, these dynamics can lead to unequal outcomes. Decisions appear neutral, yet they reflect underlying disparities.
This illustrates a key point. Data does not eliminate bias. It can encode and scale it.
Differential Visibility and the Allocation of Resources
Data also shapes how resources are allocated.
Institutions rely on data to identify needs, prioritize interventions, and evaluate outcomes. Areas or groups that are well represented in data are more likely to receive attention.
Conversely, those with limited data representation may be overlooked. This creates a feedback loop. Visibility leads to resources, and resources generate more data, increasing visibility.
Harvey (2006) connects spatial organization to power and capital. In a data driven context, this relationship is mediated through digital representation.
Inequality is thus reinforced through patterns of visibility and invisibility.
The Digital Divide Reconsidered
The concept of the digital divide has traditionally focused on access to technology.
While access remains important, inequality in a data driven society extends beyond connectivity. It includes the ability to generate, control, and interpret data.
Some individuals and institutions have the capacity to collect and analyze data, gaining insights that inform decisions. Others are primarily subjects of data collection, with limited influence over how their data is used.
Couldry and Mejias (2019) describe this as a form of data colonialism, where value is extracted from data without equitable return.
This creates asymmetries of power. Those who control data infrastructures shape outcomes, while others are positioned as data sources.
Algorithmic Decision Making and Structural Inequality
As data systems evolve, algorithms play an increasing role in decision making.
These systems classify individuals, assess risk, and generate recommendations. In many cases, they influence access to opportunities such as credit, employment, and public services.
Pasquale (2015) highlights the opacity of algorithmic systems, where decision making processes are difficult to understand or challenge.
When algorithms rely on biased or incomplete data, they can reinforce structural inequalities. Moreover, their scale allows these effects to be amplified.
The challenge is not only that algorithms can be biased, but that their influence is often hidden.
Land, Space, and Data Inequality
The relationship between data and inequality is particularly visible in land and spatial contexts.
Land governance relies on data to define boundaries, ownership, and use. However, not all forms of land tenure are equally represented.
Formal systems prioritize documented ownership, while informal or customary arrangements may be excluded. This creates disparities in recognition and access.
Data driven systems can therefore reinforce spatial inequality. Areas that are well mapped and documented receive more attention and investment, while others remain marginal.
This highlights how data shapes not only social inequality, but also spatial inequality.
A Data Justice Perspective
Addressing new forms of inequality requires a broader framework.
A data justice perspective emphasizes three dimensions.
Representation concerns who is included in data systems and how they are portrayed.
Distribution relates to how benefits and burdens are allocated.
Governance addresses who controls data infrastructures and decision making processes.
These dimensions shift the focus from technical accuracy to social impact.
Rethinking Equality in a Data Driven World
Rethinking inequality in a data driven society requires recognizing that data is not neutral.
Efforts to improve data quality, reduce bias, and expand access are important, but they are not sufficient. Structural issues must also be addressed.
This includes questioning how data is used in decision making, how systems are designed, and whose interests they serve.
Policies and practices must ensure that data driven systems do not simply reproduce existing inequalities, but contribute to more inclusive outcomes.
Conclusion
Data is transforming how inequality is produced and experienced.
It shapes visibility, influences decisions, and structures access to resources. While it offers opportunities for improvement, it also introduces new risks.
The emergence of data mediated inequality highlights the need for critical engagement. It requires moving beyond assumptions of neutrality and examining how data functions within systems of power.
Ultimately, the challenge is to ensure that data driven societies are not only efficient, but also just.
References
Barocas, S., and Selbst, A. (2016). Big Data’s Disparate Impact. California Law Review.
Bowker, G. C., and Star, S. L. (1999). Sorting Things Out. MIT Press.
Couldry, N., and Mejias, U. (2019). The Costs of Connection. Stanford University Press.
Harvey, D. (2006). Spaces of Global Capitalism. Verso.
Kitchin, R. (2014). The Data Revolution. Sage.
Pasquale, F. (2015). The Black Box Society. Harvard University Press.
samivan.id 
Leave a comment