University of Michigan - Ann Arbor

- 1,589 (71%) of participants shared at least one fake headline and moved on to the corrective measures part of the experiment and were randomly shown one of the four corrective messages designed to look like tweets with different strength (direct or hedged) and depth (simple vs. detailed)
- Replying to corrective message – “Would you reply to the above message?” – yes (1) or no (0) with free-response explanations
- Reason for correction – “Why do you think the person wrote the message you received?” with four answer choices
- Validity of the correction – participants rated accuracy and trustworthiness of the correction and author of the correction on a Likert-scale of 1-7 and whether they had positive or negative opinions of the author of the corrective message based on how helpful they found the messages
- Individual beliefs after messages – “How do you view the accuracy of the article you shared after seeing the corrective messages?” on Likert-scale of 1-5
- Cognitive reflection test (CRT) and actively open-minded thinking (AOT) tests administered to study participants’ analytical skills and openness to new opinions

You’re definitely wrong, maybe: Correction style has minimal effect on corrections of misinformation online Methodology - Corrective Measures Procedure

Martel, C., Mosleh, M., & Rand, D. G. (2021). You’re definitely wrong, maybe: Correction style has minimal effect on corrections of misinformation online. https://www.cogitatiopress.com/mediaandcommunication/article/view/3519

You’re definitely wrong, maybe: Correction style has minimal effect on corrections of misinformation online

You’re definitely wrong, maybe: Correction style has minimal effect on corrections of misinformation online Methodology - corrective measure conditions

Learn Before

Related