Seeing the forest for the trees: applying systems thinking to fact-checking

System features:

• It works as a single whole, even if it consists of many parts.
• It may have new properties that are not visible in individual elements.
• There is usually order inside: small parts combine into larger units.
• There is a constant exchange between the parts and the whole – this affects how the system behaves.
• It is rarely completely isolated, almost always exchanging resources or information with the outside world.
• Over time, the system can change and adapt to new conditions.

Fact checking is one of many areas where systems thinking, with its ability to integrate disparate data, is applied.


Applying Systems Thinking to Fact-Checking

The information that spreads in the modern world does not exist on its own. It is often embedded in the surrounding diverse contexts: political, historical, cultural, etc. The problem is aggravated by the complex media environment: information flows through dozens of channels at once, influencing completely different groups of people. In these conditions, standard methods of fact-checking can fail. It is not enough to simply analyze information – you need to be able to see the entire system of relationships. Thus, a systematic approach to fact-checking includes:

1. Identification of sources and their relationships.
Systems thinking allows you not only to check one fact, but to understand how information spreads: from the primary source to the end consumer. For example, a false news item may appear on a forum, then end up on social networks and blog platforms, increasing confirmation bias in the audience.
Analysis of connections between sources allows us to identify systemic points of distribution and nodes of disinformation.

2. Identifying patterns of recurring errors.
A systems approach helps identify patterns in misinformation. For example, a tendency to manipulate statistics, rewrite quotes or contextual fragments. Such recurring patterns can be seen as systemic “weak spots” in the media space, which facilitates the prediction and early detection of new false reports.

3. Contextualization of facts.
Fact-checking is not only about checking the truth, but also about understanding the context. Systems thinking allows you to consider the social, political, economic and cultural background of information. For example, a statement about rising food prices should be checked considering seasonality, regional characteristics and economic policy, and not just by checking the numbers.

4. Integration of tools and data.

Systems thinking helps to integrate different sources and methods: automated monitoring of social networks, network graph analysis, document verification, expert interviews. Instead of isolated checks, a holistic picture of the information environment is formed.


Network Analysis vs. System Analysis

In political communication today, fact-checking has become one of the key tools for protecting public space from manipulation. To understand its capabilities, it is important to compare network analysis and systems analysis.

The network approach shows how falsehoods circulate across accounts, media outlets, and platforms. For example, in a study of the 2016 US presidential election, it was able to identify clusters of Twitter accounts (now X) that were amplifying rumors of fraud.

A systems approach focuses on institutional and cultural conditions. For example, a Reuters Institute study found that low trust in health systems and political polarization have increased the spread of conspiracy theories about COVID-19.

Practice shows that the most productive is the use of methods together. Network analysis answers the question of “how exactly” disinformation is spread, and systemic analysis answers the question of “why” it is perceived and takes root in society.

An example of synergy between network and system analysis can be demonstrated by a study by an international team of scientists from the Universities of Zurich and Washington on the spread of disinformation during the 2022 US midterm elections.  The study took a systematic approach through large-scale quantitative analysis: the researchers processed about 446 million tweets about the election, collected in real time from September to December 2022. They examined key narratives related to the election, relying on a high-fidelity set of 135 false and misleading claims identified by academic groups monitoring social media, taking into account users’ political leanings, the virality of messages, the timing of posts, and the involvement of influencers.

Along, the researchers used a network approach to study how misinformation spreads across social media, as well as the role of influencers and users sharing fact-checking links. These links were manually collected from trusted and unbiased sites and then tracked as they spread on Twitter.

By combining systemic and network methods, it became possible to understand in detail how disinformation circulates and how verified information is disseminated. Modeling factors such as political affiliation, virality, number of influencers, and timing of posts allowed us to explain and predict the reach of fact-checks online.

This multi-layered approach demonstrates how a combination of systems analysis and network tracking can effectively combat political disinformation and create a transparent information space during elections.

Systems thinking takes fact-checking from a reactive to a proactive plane. It allows us to move beyond the question “is this statement true?” to deeper questions: “why did it appear?”, “how and who benefits from it?” and “what will make it be believed?” Research into the spread of disinformation demonstrates that the future of effective fact-checking lies in the synergy of several analytical methods, which allows us not only to combat the consequences, but to influence the causes of distortion of the information space.