searchSearch data by region...
Expert Insight

Q&A: High-Quality Data Lost Amid Noise of Fragmented Media

Politicians remain divided on how to handle the COVID-19 pandemic despite an abundance of high-quality data and analysis indicating the effectiveness of mitigation efforts such as masking. Dr. Filipe Campante attributes this to a decrease in trust in U.S. institutions and siloed social networks fueled by a plethora of biased media outlets and politicians spewing contradictory information.

Share
Authors:
Joshua E. Porterfield, PhD
October 6, 2021

The COVID-19 pandemic has increased political divisions, highlighted government paralysis, and fueled the spread of false information and biased data analyses. Dr. Filipe Campante, Bloomberg Distinguished Associate Professor of Political Economy and Governance, explains that the bad equilibrium we are currently experiencing is due to a crowded and noisy information environment brought on by fragmented, easily accessible media streams, a period of depreciating trust in U.S. institutions, and a reduction in social capital.

Why do elected officials not always lead from a purely data-driven stance?

There's a lot of information available. We're flooded with information to an unprecedented level, making it complicated to judge the quality of information. That's where political incentives emerge. Muddying that picture can play to your advantage. For example, Russia’s President Vladimir Putin realized that the state no longer needed to control information the way China does if it could flood the environment with noise. There was no need to convince people that the state was right, as long as there was no way to know what is right. Something similar has been playing out here.

“In a situation where there's a lot of noise in the information environment, data-driven policy will not necessarily be rewarded in the political market.”

The public has a hard time determining whether information is right, so they will not pay a premium for high-quality information. Also, high-quality data is harder to produce, making it costlier for the producers of information — high-quality data is hard to identify, costly, and difficult to produce. That is in the media, but it extends to governance. Even in a world where voters would like to have good, data-driven policy that leads to better outcomes, you may not get that because in a noisy information environment it's hard to identify the connections between policy and outcomes, and data-driven governance may not be rewarded.

Can you compare the role of data in governance during Ebola versus COVID-19?

The government handled it well back then. Ebola is the epidemic that didn't happen, as far as the United States is concerned. Research shows the Democrats were still punished for it, however, even though it was arguably staved off by good, data-driven policy. They lost votes in those midterms as a result of their reaction to the Ebola epidemic, because other politicians had an incentive to strategically exploit the situation. Many politicians tried to connect Ebola to President Obama and immigration, with some success with the immigration argument. The bottom line is that politicians are providing information in this environment, they have incentives to create noise, and in this case they were rewarded for it. Data-driven governance wasn't rewarded. That is still the case with COVID-19, where politicians are not necessarily being rewarded for data-driven decisions like encouraging vaccinations or mask mandates.

What changes to the information environment have enabled low-quality data to reign?

Social media plays a big role here, but it's not only that. The generally fragmented media environment allows cheaper access by politicians to media channels. Back when there were only three major TV networks, politicians were not content creators, but now they are. That is a big change that makes the information environment less conducive to a data-driven approach. It's not just the public accessing information; the politicians are players in the information environment. Politicians used to have to own a TV channel or control the state TV to flood the information environment with noise. It's easier now because access is easier.

What has been the impact of increased access to data and information on social capital?

It’s not so much the impact of data on social capital, but the other way around. Social capital boils down to the ability and the things that build the ability to engage in collective action. Trust is central to social capital because trust is what enables that collective action. In a low trust environment, the problem of the noisy information environment is magnified. You can't trust news sources if you can't trust people in general. Many people will tend to react even to high-quality information as if it’s a lie.

“This idea of social capital really gets at the heart of how you might end up in this low-quality policy-making trap because of the noisy information environment.”

The COVID-19 pandemic highlighted a lot of these dynamics. It was a new disease that we knew nothing about, so people had to rely on external information. It is hard to figure out who actually knows what they're talking about because we are in a low-trust environment. The notion of social capital is critical here because it helps determine whether we're going to be in that bad equilibrium where nobody knows what's going on because you can’t trust anything. You can contrast the response that happened here, where we're living through a period of depreciating social capital, versus a place like Germany. There the majority sentiment is that the government generally knows what it’s doing and the public can follow its lead.

How do we cut through the noise to support high-quality data and governance?

It’s hard. One approach from early in the pandemic was fact-checking. If we can’t prevent low-quality information from getting out, we can add another piece of high-quality information to refute it. But it's not always the case that good information will drive out bad information, it can also be the reverse. Fact-checking has often been ineffective because it just adds more information to this already overcrowded information environment. Oftentimes the way people process high-quality fact-checking information will be through that noisy lens, and they’ll say if the fact-checking goes against their views, it’s just "fake news."

Quality is hard to observe. People are going to choose their data sources through trust. Therefore, we need to look for trust relationships to share high-quality information. For instance, if we have a community that is reluctant to get vaccinated, we need to find a messenger that the community trusts. Trust is of the essence here because we're in a situation where it's hard to assess the quality of information. We need to find trustworthy messengers, knowing that there are incentives on the other side to fight back with noise.

Joshua E. Porterfield, PhD

Dr. Joshua E. Porterfield, Pandemic Data Initiative content lead, is a writer with the Centers for Civic Impact. He is using his PhD in Chemical and Biomolecular Engineering to give an informed perspective on public health data issues.