Hello lovely coursemates! It was so nice to see you all digitally and physically during the Inclusive Society intensive. The valuable contributions you all made have really helped me to shape my thoughts on what I want to explore in my final project.

The below is an exploration of some principles that I’ve been slowly developing on the theme of community approaches to data. It’s part of some work that I’m putting together with the aim that it might inform my work at charity Community Energy Scotland where we work with communities who have data they’d like to use, or are being requested to share beyond their borders. The agency that data creates is something I’d really like to explore more during my final project.

I’d love to hear your thoughts on any of the points below. Please do share any references/ thoughts/ criticisms.

Replacing “datafication” with activism 

We live in an era of unrelenting quantification. Many decisions made in our society are the result of vast amounts of data extracted, often without knowledge or explicit consent, from the population. We recognise that data has “transformative and affective potential” (Renzi and Langlois, 2020). Communities “making-things-into-data can be an exercise of agency, rather than merely succumbing to external pressure.” (Blacker, 2021).  

In data sharing projects, how can we ensure that the providers of the data (not just the data controller) are given something positive in return for their contribution rather than just being extracted from? How can we ensure their context is kept with the data in future uses? How can we be accountable for, and report back on the impact their data has had? 

Measurement and “Situated Knowledge” 

We acknowledge that the process of gathering data is not a neutral act. The data itself fundamentally reflects the context (social, political, locational) of those collecting the data as well as the collectors’ stance and assumptions (Blacker, 2021). This means that data cannot be seen as a neutral entity. This must be considered when any decisions are made about the sharing and use of that data. In sharing data, it must be acknowledged that this context can easily be lost (Blacker, 2021). 

Shawn Wilson (quoted in Liboiron, 2021): “The problem with that is that we can never really remove the tools from their underlying beliefs. Since these beliefs are not compatible with our own, we will always face problems in trying to adapt dominant tools to our use.” How can we consider the shape and form of data sharing tools themselves in working for the objectives of margianlised groups? 

Sharing and Formatting 

Blacker describes that not all data must be formatted numerically to carry out the function intended. Instead, other forms such as narrative form can be used to communicate the message (2021). Alternatively, summary datasets can be used that incorporate local knowledge and context, as well as obscuring data in a way that makes it only applicable for certain contexts. 

How can our work ensure such metadata and supplementary context data (e.g. narrative data, summary data, augmented data) is shared in tandem with the original dataset? 

Refusal 

Communities need only share their knowledge, and data just enough to achieve the aims that benefit them. This approach stems from an approach used by indigenous communities in Canada fighting pollution. The level of data being shared should allow for the impacts of the pollution to be communicated in a language that can be comprehended by legislators; without allowing for the assimilation of traditional knowledge which may be used for other purposes (Blacker, 2021) as such ongoing loss of data sovereignty can exacerbate existing inequalities. 

The refusal of access does not necessarily mean that nothing will be offered. In its place the data subjects or community can help the requestor to understand the refusal and direct them to resources that might re-align vlues (Liboiron, 2021). 

Data’s Fundamental Relationality 

Viljoen (2021) recognises that data does not just impact those who it directly gives insight on. Data can be used to draw inferences about other people who share characteristics in common with the observed group. This  

means that all data governance must consider the potential harms to other people who share social groups. 

Goal 

The goal of data processes is that the community themselves would be able to carry out their own research using these methods (Liboiron, 2021). All other forms of data sharing are to support the community, on their terms, in the absence of this possibility.   

References 

Blacker, Sarah. “Strategic Translation: Pollution, Data, and Indigenous Traditional Knowledge.” The Journal of the Royal Anthropological Institute 27.S1 (2021): 142–158. Web. 

Liboiron, M. (2021) Pollution Is Colonialism. Durham, NC: Duke University Press. 

Renzi, A. and Langlois, G. (2020) ‘Renzi, A. and Langlois, G. (2015) “Data / Activism” Elmer, G. et al. (eds.)Compromised Data: From social media to Big Data, , London: Bloomsbury, pp. 202–225.’, in, pp. 202–225. 

Viljoen, Salome. “A Relational Theory of Data Governance.” The Yale law journal 131.2 (2021): 573–654. Print. 

Icons are CC from thenounproject. Individuals credits are embedded on the images.

Community Data Practices / Knowledge Integration and Project Planning: Data, Inequality and Society (2022-2023) by blogadmin is licensed under a Creative Commons Attribution CC BY 3.0