Who isn’t Counted? Assessing Inclusion Challenges in Data Collection and their Implications for Good Governance — PART 2

Governments, civil society, and concerned citizens are increasingly turning to data for smarter governance. Most data-driven practices focus on what data exists to decide which governance issues to address. The 100 Questions Governance Domain reinvents this process by asking what are the most pressing questions, as determined by experts and the public, and then takes steps to find and collect the data that can help solve these issues.

Last month, The GovLab hosted two online panels with experts from multiple sectors and regions to address these issues within the context of the 100 Questions Governance Domain.

Our first panel, “Building Data Governance Strategies,” emphasized the importance of fostering a governance culture that embraces data use and provides resources to maintain data collection and analysis. As well, the panelists discussed sectoral silos, which prevent information from being shared between government agencies, private corporations, and civil society that could create mutual benefit by engaging with constituents and increasing decision-making transparency and legitimacy.

Following this session, we hosted our second panel, “Doing Data-Driven Democracy Right.” This article recaps that conversation below.

Read more about “Building Data Governance Strategies” in Part 1 here and find our “Cross-Cutting Takeaways from the Panels” in Part 3 here.

Panel 2: Doing Data-Driven Democracy Right

Moderated by The GovLab’s Knowledge Director, Andrew Young, this panel featured:

  • Renata Ávila, CEO at the Open Knowledge Foundation;
  • Rosario Pavese, Senior Regional Coordinator, Americas of the Open Government Partnership; and
  • Silvana Fumega, Project Director of the Global Data Barometer & Research and Policy Director at the Latin American Initiative for Open Data (ILDA).

Following a short introduction to the 100 Questions Initiative and the Governance Domain, Andrew and the guests discussed avenues of using data to improve governance. Bringing a Latin American perspective to the conversation, Renata, Rosario, and Silvana emphasized that governance culture greatly influences data-sharing infrastructures and underscored the importance of safeguarding vulnerable populations before implementing data collection.

The full 90-minute talk, as well as an overview of the main themes, is available below:

The Role of Data in Governmental Decision-Making

Andrew kicked off the panel by asking all the guests about why more effective uses of data are important for governance. Renata started the discussion, pointing out the challenges of nepotism, corruption, and opaque decision-making that plague local and national governments in Guatemala. These practices reinforce assumptions about people and situations on the basis of feelings rather than facts, ultimately doing constituents a disservice by not illuminating the root cause of issues. “A good data set makes visible the invisible,” she noted, emphasizing the need for information to understand why certain issues exist in order to take a better approach to problem-solving.

Silvana added that even though data has become more public, how the information is used remains unclear. Scoping the mechanisms by which data is employed to address social problems is a key focus of the Open Data Barometer project. While there exists a correlation between data availability, governance, and use and impact, she noted that the operationalization of data is low across nations. This signals a gap in government abilities to apply the data they have to the issues at hand.

Rounding out the discussion, Rosario echoed Renata and Silvana, stating that the current challenge is not data availability but rather efficient data use. She reinforced the need for data use to be grounded in human rights to solidify the connection between transparent information and democratic practices.

The Current State of Data Policy

Considering the gaps in knowledge, infrastructure, and human capital in the current data policy environment, Renata noted that many governments do not produce data themselves because they do not consider data as an important tool for policymaking. “There is a system that gives certain privileges for the private sector […] that are not accessible to universities and organizations that are working in the public sector.” Moreover, austerity measures force Central American governments to stick with analog systems, causing public bodies to fall behind the curve and to rely on private-sector data, which can cause issues of data provenance and availability. By not embracing data as a key pillar for government infrastructure, regulators experience missed opportunities to use data.

The Ethical Concerns Around Data-Driven Governance

Turning to Silvana, Andrew asked her about the role that data responsibility, rights, biases, and ethical concerns play in data-driven decision-making.

She pointed out the gaps in the data that we already have, which do “not usually count all people and all contexts.” Gathering the right data requires extensive forethought on all the variables needed. When addressing crime in Central America, Silvana notes that multiple datasets — from femicide to migration patterns — are brought together to understand the whole context of where crime occurs, and against whom. However, narrow scopes in the collection of these data omit some individuals. For example, datasets that only consider binary gender categorizations exclude individuals that identify in other manners, and subsequently miss crimes and violence against them, creating a policy gap that marginalizes and endangers these members of society.

“What is not counted is not seen, and if it is not seen it is very difficult to act on it.”

Andrew later followed up on the issue of data visibility, particularly with how it relates to communities that might want to remain data invisible. Silvana recounted her experiences with this concern in migration work and described the need to standardize and anonymize data to avoid the identification of people, especially those who are part of small and/or vulnerable groups. Renata added the need for cross-sectoral data literacy. “I think the people working in data need to understand more about society and politics, and I think that people working in politics … need to gain data literacy.”

Responding to audience questions about how policymakers should think about data literacy training and introduce behavioral science tools to better understand open data, Silvana highlighted the role of intermediaries, such as data stewards, to bridge the knowledge gaps between data and analysis. Rosario concluded by saying that we need to find “synergy between governments and society” to discuss and use data in a mutually beneficial manner.

The Role of External Actors in Shaping Governments’ Data Use

Next, Rosario talked about how other actors, specifically, non-profits, watchdog organizations, and citizens groups, influence the way governments use data. “We have seen that civil society has been the determinant in influencing all the open data and open government development[s] … particularly, the open data community in Latin America is luckily very vibrant and challenging [of governments]. We have seen a huge effort in society in promoting these types of commitments.”

Between 2011 and 2014, the Open Government Partnership found that open data commitments focused on high-level, general objectives, such as creating open data portals and repositories and establishing national open data policies. In 2015, efforts to open data became more precise: by targeting specific sectors or seeking data for specific purposes, external players started to enter the mix. Now, an overarching goal is to bring in those who have not been historically considered in the open data movement to cater to their specific needs and interests. The widening of audiences and interests introduces new considerations on how governments gather and use data.

The Panelists’ Most Important Question from the Top Ten Governance Domain List

Andrew then presented the top ten questions of the Governance Domain, which were curated by the bilingual community and are now available for public voting. He asked the panelists which one question was the most pressing to each of them.

Rosario chose “Does open governance lead to more inclusive, transparent and timely decision-making?” because of its relevance to the work undertaken by the Open Government Partnership.

Silvana emphasized the importance of understanding who is and is not counted, selecting “Which groups are included and excluded in the data used for formal government decision-making?” as her top question.

Renata mentioned that all of the top ten questions are interconnected, and echoed Silvana’s choice. “If we get it wrong here [at the inclusion and exclusion stage], we are going to get wrong what comes next and what comes next.”

***

About The 100 Questions Governance Domain

In 2021, The GovLab, along with partners from The Asia Foundation, the Centre for Strategic and International Studies, Indonesia, and the BRAC Institute of Governance and Development, launched the Governance Domain. This domain marks the sixth installment of the 100 Questions Initiative, a cross-sectoral project that seeks to collaboratively map the 100 most pressing, high-impact questions. Unlike traditional problem-solving methods, which formulate questions based on the data already available, the 100 Questions challenges “bilinguals’’ — a global cohort of subject matter and data science experts from across industries — and the public to focus on defining the problem first, then seeking out the data that could help address the issues. By taking a demand-driven approach to policymaking, this approach identifies which issues are the most urgent and lays the foundation to build data collaboratives, a new form of public-private partnerships that harness intersectoral data for public good.

--

--

--

Responsible Data Leadership to Address the Challenges of the 21st Century

Recommended from Medium

Case Study: Efficient Daily Scrum with Silent Meetings

5 Advanced Visualisation for Exploratory data analysis (EDA)

Predictive Modeling with Iowa State Liquor Sales Data

Visualizing Linear, Ridge, and Lasso Regression Performance

Analysis of an Article

Analyzing the Utility of Ichimoku Analysis

Training a Computer Vision Model on VertexAI (with and without Explanable option)

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Uma Kalkar

Uma Kalkar

More from Medium

Pursuing Purpose & Service with Mark Voss

Movement building and the climate emergency

Lessons from Vik Bakhru, Circulo: building a new Medicaid model centered around the patient

macOS Rootkit Emulation

macOS Rootkit Emulation