Influence isn't about data, it's about compelling stories. But when Congress uses stories, where are they getting the data to support them? Are they just anecdotes or is there evidence behind them? Will big data and sophisticated data analytics change how these stories are told? After attending two events related to data and Congress, I've determined that the rise in open data and simplistic data visualization will make data even easier to misunderstand and manipulate than it already is. To ensure this does not happen, Congress will need to place significant thought and attention into how they are obtaining, interpreting, and using data.
Recently, I attended two events related to Congress and data: the Legislative Branch Capacity Working Group session on Evidence-Based Policy-Making and the DATA Act Summit. As I discovered at these two events, the rise of open data and data analytics are going to change how Congress obtains, interprets, and uses data to tell stories and make decisions. Ensuring the changes are good for democracy will require significant thought and attention by Congress.
During the Evidence-Based Policy Making event, panelists discussed the challenges Congress faces when implementing evidence-based policy making. Dr. Andrew Reamer, a public policy Research Professor at George Washington University highlighted the value of stories within policy making. He emphasized that "public policy is all about storytelling and competing stories." Legislators need persuasive rhetoric to win support. Interpreting data so it tells a compelling story can change policy. Here at CMF, we know how important storytelling is for the policymakers on the Hill. But how reliable is the data and the interpretation, and how do Members and staff differentiate good data from bad?
According to Timothy Shaw of the Bipartisan Policy Center, the current sources Congress relies on for data aren't that great, and "[Congress] currently has limited capacity for trusted sources of information." This issue was also raised by Representative Will Hurd (TX-23) at the Legislative Data and Transparency Conference. Hurd said, "We don't even know what questions to ask because we don't have the right data." But new legislation like the Digital Accountability and Transparency (DATA) Act is trying to help.
The DATA Act is the first federal open data law in the U.S. The act aims to make information on federal expenditures public, accessible, and transparent by standardizing how all federal agencies report spending. If everyone uses the same structure and terminology, the data is easier to understand and analyze, and if it's publicly available the analysis can be performed by anyone. As a result of the DATA Act, Congress and citizens will, for the first time, be able to analyze federal-wide spending in detail and in one centralized place. That's a big deal. As Rep. Hurd said, "The federal government has been around for how long, and we don't have all our financial data in one place that we can gain access to?"
The centralization of all federal spending data is great news, especially for Congress, as the body responsible for making decisions about federal spending data. But as a researcher, I can't help but ask the larger question: What does this all mean? Although I'm excited about the prospects for easy access to federal spending information, there are potentially critical downsides which need to be discussed. On June 29th, the Data Coalition hosted the DATA Act Summit to celebrate the implementation of the DATA Act, and to reflect on the work which has been completed thus far. As the DATA Act Summit demonstrated, an entire market has sprouted up around open access to this data, and dozens of companies are creating software for visualizing federal spending data. How this data is visualized and interpreted--and by whom-- will impact what stories are told and how public policy decisions are made.
My concern is that the rise in open data and simplistic data visualization will make data even easier to misunderstand and manipulate than it already is. Think fake news, which almost always includes real data. Data visualizations tend to focus on raw numbers, which are easy to manipulate into different stories depending on a person's needs. Raw federal spending data is no different. If a congressperson sees a graph that says the government spends more money on socks than ties, what does that really mean? Are socks more expensive? Do more people wear socks than ties? Are socks mission-critical to an agency that's actually being very frugal with its sock purchases? Without more information, that small piece of spending data can be interpreted however a person likes. Yet, developing data visualizations that I saw at the DATA Act Summit aren't acknowledging that context. The story behind the numbers is missing.
During the summit, I expressed my concerns to almost every company and agency that presented demos of their federal spending visualizations. I asked what they are doing to help users understand the story behind the numbers. Most aren't doing much. They assured me that their visualizations weren't meant to be conclusions, but to be the starting point for further investigation. But when and by whom? Will a busy congressional staffer continue to investigate past the numbers? Will a citizen or lobbyist? Although I am excited about the prospect of open information on government spending, I fear the lack of context will lead to messy and misguided interpretations of the data.
All data has a story, but that story can be misinterpreted or manipulated without contextual information. When Congress makes decisions based on wrong or manipulated stories, the result is legislation and ultimately law that doesn't address the whole problem or, worse, addresses the wrong problem or something that isn't a problem at all. At the Evidence-Based Policy Making event, Reamer stated, "Congressional staff need to know how to be good consumers of evidence." I strongly agree. We need Congress to thoroughly understand how to think critically about data and the interpretations and visualizations based on them and to ensure that data access and visualizations provide not only the raw numbers, but also the story behind them.
Samantha McDonald is a summer research assistant at CMF. She is also a PhD student at the University of California, Irvine studying the effects of information and communications technology on the relationships between policymakers and citizens.
