Data Governance: the Human Decisions that Drive Data and Technology
Written by the DSH Coordination Team* in collaboration with Samuel Kohn, Program Director at Open North
Une version française de ce billet a été publiée ici.
This is the first article in a collaborative series between the Data for Society Hub (DSH) and Open North (ON), Montréal in Common’s data governance project leader. As such, the ON team is assisting the DSH team in its efforts to design and develop collaborative, consistent, and robust data governance frameworks. This series of blog posts will discuss the relationship between governance and technology, the specific needs of the DSH partners, and the implementation of the chosen framework.
It has been said before: data pooling involves significant inter-organizational, ethical, and legal considerations. Consequently, the solution developed by the Data for Society Hub (DSH) must be backed by a data governance framework, which will guide the technical and human aspects of data usage and management.
Data, Technology and the DSH
“Data may be used to obtain information, draw conclusions, and inform individual or group decision-making. Information and communication technologies enable the collection, hosting, transformation, and sharing of this data. Data and technology go hand in hand, but they must first and foremost meet concrete needs,” says Samuel Kohn, Program Director at Open North (ON). He notes that data are constructed with intent that is not neutral: one or multiple individuals make decisions about what the data represent and how they are utilized [1].
The technological solution developed by the DSH aims to pool together data on Montréal from various sectors. We seek to enhance the value of these data through the individual or collective projects of partner organizations in order to contribute to the well-being of the population. Our success depends on several factors. Samuel pinpoints two that are data-related: “the first is building stakeholder confidence in the proposed tools for decision-making and responsible use of data; the second is ensuring that the data is actionable.”
Simply put, users must be able to quickly access the information they need and determine whether the data is of sufficient quality for their intended purpose. They also need to know the context in which the data were collected, their limitations, and the transformations they have undergone through the various stages of the data life cycle (see diagram below) to ensure their proper interpretation and use. This is where data governance comes in.
The Role of Data Governance
Data governance “encompasses all the factors that influence decisions about data (e.g., whether or not to acquire the data; how to consider the different needs of the population, directly involving them or not; and chosen techniques and technologies for data collection, storage, processing, and analysis, as well as their use),” reads Open North’s Introductory Approaches and Tools for Data Governance [2].
At the DSH, the data governance framework has three purposes. First, it allows participating partners to make explicit the strategic and ethical objectives of data usage, i.e., to gain insight into the specific issues that affect Montréal’s population. Second, it aims to outline the data policies, guidelines, and procedures in place to achieve this goal. Finally, it helps establish a relationship of trust with the public.
What are the main criteria for effective governance? “For the Montréal in Common (MIC) project, ON advocates for accountable, efficient, and collaborative data governance.This means that decisions are made based on the risks and potential benefits to stakeholders, in addition to aiming for useful outcomes. This also means that decision-makers are continually seeking to improve their management practices through the practice of knowledge-sharing,” says Samuel.
We want our data governance framework to be leveraged for the common good within and through the DSH [3]. There are also numerous internal and external factors to be considered in our project:
- The way the organization or group operates;
- Existing priorities and strategies;
- Corporate culture;
- The type of data involved;
- The legal context;
- The norms and standards of the work environment.
The governance framework must adapt to the project’s particularities and evolve over time. According to our collaborator, this “requires constant effort, both to monitor the many impact factors and to involve stakeholders in the processes that concern them, in order to understand their evolving needs and expectations. There are multiple long-term benefits for the DSH: greater relevance of data decisions and processes; stakeholder engagement regarding changes; contributions to the project’s sustainability; etc.”
Our Data Partnership
The DSH will be responsible for the design and implementation of a data governance framework as a system of standards and rules for decision-making around sharing and using data. This framework will be informed by the key concerns of the project partners and their beneficiaries, and will be consistent with the principles outlined by the City of Montréal in its Digital Data Charter (2020).
Indeed, this is one of ON’s main recommendations in a report produced for Montréal in Common: “Given that the DSH strives for data valorization that considers the welfare of Montréal’s population, its data governance framework should strongly address ethical principles of data.” [4] Consider privacy rights, transparency, universal access, etc.
Our governance framework will bridge the gap between the technical side of the project (which involves the development of a digital infrastructure for data pooling, visualization, and analysis), and critical human aspects such as social development, communication, collaboration among stakeholders, and the harmonization of practices.
DSH stakeholders include data holders (partner organizations), potential users (the academic, public, or community sectors), governing bodies (the project coordination team and MIC), regulators (governments and institutions) and beneficiaries (the Montréal population). Today, not all stakeholders have an equal ability to influence dataset negotiations — hence the importance of including structurally underrepresented groups [5] through fieldwork.
Human Needs and Everyday Practices
Data governance already exists implicitly in our project. The current challenge is to develop conscious and coherent mechanisms in an inclusive and participatory manner based on the following elements:
- The partnership’s strategic objectives;
- Current stakeholder data decisions;
- The internal and external factors that underlie them.
“Open North has already developed a data governance framework for the MIC innovation community that documents concrete actions to operationalize the principles of the Digital Data Charter,” says Samuel. This is a common baseline that can inform the DSH as an MIC project. It will be possible to tailor this framework: workshops, discussion groups, experiments or tests to address specific use cases [6] may provide learning opportunities. These, in turn, can influence certain technological choices that may facilitate or automate some of the mechanisms.”
By the end, we will have a clear idea of the DSH’s starting point, its desired trajectory, and the conditions necessary to achieve this vision. These first two components will be the focus of the next feature in this series: how do partner organizations currently manage their data, and what needs are they seeking to address through our data pooling project? A final blog post will then focus on implementation tactics for the governance framework.
[1] For background information on the nature and governance of digital data, see this post (in French) by Lauriane Gorce, Program Manager at ON.
[2] Open North, Introductory Approaches and Tools for Data Governance: Towards Accountable, Efficient, and Collaborative Data Governance, Montréal in Common, January 2022.
[3] The use of data for public welfare involves mobilizing it to build a more just society, overriding short-term considerations and the private interests of organizations or individuals (see Open North, Montréal in Common’s Data Governance Framework. Towards Accountable, Efficient, and Collaborative Data Governance, Montréal, Montréal in Common, January 2022).
[4] Open North, Experiments in Collaborative Data Governance: Where to Begin? Recommendations for the Data for Society Hub, Montréal, Montréal in Common, September 2021.
[5] There are several levels of under-representation: for example, under-representation of certain communities in the data itself (as illustrated in this example) or in decisions about the data where these communities will be most affected (at the center of data governance). In our case, the challenge is to demonstrate the extent to which we will allow Montréal’s various communities to participate in decisions related to the data that concerns them.
[6] A use case refers to “[m]any ways of using data that have value or utility for the actors involved. A data use case corresponds to a well-defined problem, in a specific context, as well as to a set of actions carried out by the actors and stakeholders of the data in question in order to achieve an objective, a purpose” (Open North, Montréal in Common’s Data Governance Framework. Towards Accountable, Efficient, and Collaborative Data Governance, Montréal, Montréal in Common, January 2022).
Editorial: Julie Levasseur and Samuel Kohn (NO); content editing: Karolyne Arseneault, Lauriane Gorce (ON) and Julien Vallières.
The Data for Society Hub is a project by Montréal in Common, a community for the development of innovative projects for the Smart Cities Challenge.
