Debunking misleading graphs effectively: How vocationally educated young adults perceive graphs

PLoS ONE, vol. 21, pp. e0340100

Abstract

Misleading graphs can give readers a distorted view of the underlying data. We want to know how to most effectively correct misleading graphs and if it matters whether a correction uses the full-design of the original or a clean design with all embellishment and colors removed. We focus on vocationally educated young adults, a group that is vulnerable to misinformation and has so far been underrepresented in research. We use a mixed-method approach with a qualitative think-aloud task (n = 10, data collected in April 2023) and a quantitative survey (n = 130, data collected between July and October 2023). The think-aloud task showed that vocational students use a combination of calculating and estimating to process graphs, which contradicts existing literature, and that their perception is heavily influenced by context. The survey showed that graph corrections work in reducing misleading effects and also have a learning effect such that students are less misled by new misleading graphs of the same type. There was no difference between full-design and clean design corrections. These results imply that vocationally educated young adults can benefit from seeing corrections of misleading graphs.

Authors 4

1. Winnifred Wijnker

   University of Applied Sciences Utrecht

   Affiliation as printed

   Research groups Qualitative Journalism in Digital Transition and Human Experience & Media Design, Utrecht University of Applied Sciences, Utrecht, the Netherlands

2. J.P. Burger Aachen Journalism and New Media

   Leiden University

   Affiliation as printed

   Journalism & New Media, Leiden University, Leiden, the Netherlands

3. Ionica Smeets Aachen Science Communication and Society

   Leiden University

   Affiliation as printed

   Science Communication & Society, Leiden University, Leiden, the Netherlands

4. Sanne J. W. Willems corresponding

   Amsterdam UMC Location University of Amsterdam

   Affiliation as printed

   Department of Epidemiology and Data Science, Amsterdam UMC, location University of Amsterdam, Amsterdam, the Netherlands