Uncertainty words are prevalent in misinformation posts. They can make a post look more trustworthy and relatable, and therefore more acceptable to share. This could be because sharing the uncertain post does not force the user to accept responsibility, should the claim be false. 
- Examples: someone, sometimes, possibly, appears to be, others may say, etc.  

Linguistic Characteristics of Misinformation: Uncertainty Words

Emotion words can affect a person's emotional state by making them more aroused or evoking a certain emotion. Misinformation creators tend to use more emotion words (specifically negative emotion words) to elicit an emotional response from users. This can have an impact on their social media behavior, such as their likelihood of liking or sharing a piece of misinformation. 

Linguistic Characteristics of Misinformation: Emotion Words

Persuasive words are used to change a person's attitude or behavior. 
-Examples: An expert says, studies have shown, A specialist in the field believes, according to a trusted scientist, etc. 

Linguistic Characteristics of Misinformation: Persuasive Words

Misinformation creators may use comparative words in their content to make themselves appear more reliable. They may use a similar situation to imply they have experience with the topic they write about. 

Linguistic Characteristics of Misinformation: Comparative Words

Previous research has shown that misinformation tends to contain more uncertainty words and be more emotionally extreme. What is unknown is the impact these characteristics have on misinformation dissemination. To address this gap, Zhou et al. examined four linguistic characteristics thought to have an effect on misinformation dissemination, including:
- Uncertainty words
- Emotion words (positive and negative)
- Persuasive words
- Comparative words

University of Michigan - Ann Arbor

According to a study conducted by Cheng Zhou, Kai Li, and Yanhong Lu, misinformation containing specific linguistic qualities is more likely to be shared on social media. By examining the content of fake news posts and the number of shares, they found that misinformation containing persuasive and uncertainty words were more likely to be shared than true posts. The effect was moderated by the richness of the information. 

Linguistic Traits of Misinformation Online

Zhou, C., Li, K., & Lu, Y. (2021). Linguistic characteristics and the dissemination of misinformation in social media: The moderating effect of information richness. Information Processing and Management, 58(6). https://doi.org/10.1016/j.ipm.2021.102679

Linguistic characteristics and the dissemination of misinformation in social media: The moderating effect of information richness

Proposed by Daft and Lengel in 1986, the theory defines media richness as the information load of the content. For example, a plain text post is less rich than a post with an image. 

Learn Before

Related

Learn After