~/talk

The Untruthful Art - Five Ways of Misrepresenting Data

Talkat NDC London 2021 · 2021-01-25..01-29· [data-visualization, tech-ethics]

speaker

name:Alexander Arvidsson

linkedin:alexanderarvidsson

resources

video:https://youtu.be/hOeKlz4owzY

slides:none on file

summary

Five concrete tricks for misrepresenting data: axis manipulation (National Review and Fox News charts that hide trends by rescaling or removing axes, Excel/Power BI defaults make this easy), cherry-picking (selective years or respondents to manufacture trends like the Sweden crime index or UK debt slices), faulty comparisons (maps and cumulative vs. absolute revenue), and correlation≠causation (MMR, glyphosate examples) leading up to overt purposeful bias in political graphics (BJP petrol chart, altered UK COVID scales).

AI-written summary · about

chapters (5)

related talks