How to Spot a Misleading Environmental Graph

Current figure

There is no single global figure for the prevalence of misleading environmental graphs, as no systematic monitoring exists. However, a 2018 analysis by Carbon Brief of 200 graphs in UK newspaper articles about climate change found that 15% employed at least one technique that could mislead readers, such as a truncated y-axis or a cherry-picked time period (Carbon Brief, 2018). This figure provides a snapshot of the issue in one country’s print media.

Measurement date

The Carbon Brief analysis was published in 2018 and examined graphs from UK newspapers over a 12-month period in 2017–2018. The study is not updated regularly; it was a one-off investigation. No equivalent global or annual survey exists.

Previous figure

A 2015 report by the Union of Concerned Scientists, “Is the Media Betraying the Public on Climate Change?”, analyzed graphs in climate contrarian blogs and found that 20% contained misleading elements (Union of Concerned Scientists, 2015). The two studies are not directly comparable due to different media types and methodologies, but they suggest that misleading environmental graphs have been a persistent concern.

Long-term trend

No consistent long-term data tracks the frequency of misleading environmental graphs. Anecdotal evidence and media criticism suggest that awareness of graphical integrity has grown since the 1980s, spurred by works like Edward Tufte’s “The Visual Display of Quantitative Information” (1983). However, the rise of digital media and social sharing may have increased the spread of poorly designed or deliberately deceptive charts. The following table summarizes key milestones in the study of graphical misinformation:

Data source

The primary data sources for the prevalence figures cited are:

• Carbon Brief, “Analysis: How UK newspapers cover climate change”, 2018. Available at carbonbrief.org.
• Union of Concerned Scientists, “Is the Media Betraying the Public on Climate Change?”, 2015. Available at ucsusa.org.

General principles of graphical integrity are drawn from Tufte (1983) and Cairo (2019).

Methodology

Both the Carbon Brief and Union of Concerned Scientists studies used content analysis. Researchers collected a sample of graphs from specified media sources, then coded each graph for the presence of predefined misleading techniques, such as:

• Truncated y-axis (not starting at zero when appropriate)
• Cherry-picked time periods that exaggerate or hide trends
• Omitted baseline or reference line
• Inconsistent scales on dual axes
• 3D effects that distort proportions

Inter-coder reliability checks were performed to ensure consistency. The percentages represent the share of graphs in the sample that contained at least one such technique.

Why annual values fluctuate

Because no regular survey exists, there are no annual fluctuations to report. However, the prevalence of misleading graphs in environmental reporting can vary from year to year due to several factors:

• Media attention cycles: During major climate events (e.g., COP summits, extreme weather), the volume of graphs increases, and rushed production may lead to more errors.
• Editorial standards: Newsroom training and guidelines on data visualization can improve or deteriorate over time.
• Political pressure: In periods of intense climate debate, advocates on all sides may produce more deliberately misleading charts.
• Technological change: New tools make it easier to create sophisticated but potentially deceptive visualizations.

Regional variation

No comprehensive global comparison exists, but studies suggest variation by country and media type. The 2018 Carbon Brief study focused on UK newspapers; a 2021 study by the George Mason University Center for Climate Change Communication (unpublished pilot) found that 25% of climate graphs in US newspapers contained misleading elements, though the sample was small. The table below summarizes available regional snapshots:

These figures are not directly comparable due to different methodologies and sample sizes.

Meaning and limitations

The prevalence figures indicate that a notable minority of environmental graphs in news media may be misleading, but they have important limitations:

• Sample bias: Studies often focus on specific outlets or topics, not all environmental graphs.
• Subjectivity: Determining whether a graph is “misleading” can involve judgment; some techniques (e.g., truncated axes) are not always inappropriate if clearly labeled.
• Impact not measured: A misleading graph that is widely shared may have a greater effect than many accurate ones, but prevalence studies do not capture reach or influence.
• No trend data: Without regular surveys, it is impossible to say whether the problem is getting better or worse.

Readers should use these figures as a rough indication of the issue, not as a precise measure.

Next expected update

There is no scheduled update for the Carbon Brief or Union of Concerned Scientists analyses. The George Mason University study is expected to be published in a peer-reviewed journal in 2024, but no date has been confirmed. In the absence of regular monitoring, media watchdog groups and academic researchers occasionally release new studies.

Downloadable chart or table

The table below lists common techniques used in misleading environmental graphs, along with tips for spotting them. This is not based on a single dataset but represents a synthesis of guidance from Tufte (1983), Cairo (2019), and the Data Visualization Society.

Readers can download the underlying datasets from the original sources: Carbon Brief’s analysis is available at carbonbrief.org, and the Union of Concerned Scientists report at ucsusa.org. For general guidance, the Data Visualization Society’s “Best Practices” guide is freely available online.