A recent study published in Communications Psychology sheds light on the potential psychological impacts of the social media website commonly known as Twitter (now known as “X”). The findings, based on experience sampling methods to capture real-time data, suggest that the use of Twitter is associated with decreases in well-being, and increases in political polarization, outrage, and sense of belonging.
Despite Twitter’s relatively smaller size compared to platforms like Facebook and TikTok, its significant role in shaping public opinion — especially among elites in entertainment, journalism, and politics — makes it a critical area of study.
Previous studies have linked social media use with political polarization, expressions of outrage, and declines in subjective well-being. However, these studies often rely on public Twitter data, which may not accurately reflect the average user’s experience. This study aims to bridge that gap by examining Twitter’s impact using a sample more representative of the general population and by focusing on the platform’s specific features.
“My fascination with social media platforms began in my teenage years, drawn by the allure of virtual environments designed to simulate social interactions,” explained study author Victoria Oldemburgo de Mello (@vicoldemburgo), a PhD student at the University of Toronto.
“This interest deepened with Twitter, a platform distinguished by its unique dynamics. Despite not being among the largest of its kind, Twitter is a hub for highly influential individuals. It has a distinctive way of promoting certain types of content and behavior, which appears to foster an environment where expressions of outrage are more common.”
To dissect the psychological impacts of Twitter use, the researchers used Prolific Academic to recruit a sample of 252 individuals who used Twitter at least twice a week. This group was more representative of the broader Twitter user base in terms of age, gender, and race compared to previous studies, which often relied on convenience samples like undergraduate students. Data was collected between March and June 2021.
The core of the study involved sending participants surveys five times a day for a week, between 9 a.m. and 10 p.m. This design, known as experience sampling, allowed for the capture of participants’ experiences and reactions in real-time, ensuring that the data reflected genuine, spontaneous interactions with Twitter. This method reduces the potential for recall bias and provides a more accurate understanding of the psychological impact of social media use in daily life.
In each survey, participants reported whether they had used Twitter in the preceding 30 minutes and, if so, detailed their activities on the platform. These activities were categorized into observable behaviors, such as scrolling, liking, tweeting, and messaging, as well as the functions of these behaviors, like seeking information or entertainment.
Well-being was gauged using a modified version of the Scale of Positive and Negative Experience, asking participants to rate their feelings over the past 30 minutes. This tool helped quantify momentary well-being as a balance between positive and negative emotions.
Additionally, participants’ sense of belonging was measured through a two-item scale, and political polarization was assessed using a thermometer scale to gauge warmth towards Democrats and Republicans. Outrage was operationalized as a combination of anger, disgust, and repulsion.
The data indicated that Twitter use, on average, was associated with lower well-being. Specifically, when participants used Twitter, they reported a decrease in well-being to the tune of 0.10 standard deviations. This effect was found at the within-person level, suggesting that Twitter use could momentarily dampen users’ mood. Additionally, the study found Twitter use to be associated with increased feelings of boredom and loneliness, particularly among frequent users.
Contrasting its effects on well-being, Twitter use was positively related to a sense of belonging. This finding indicates that interactions on the platform can foster a feeling of community and connectedness among users. The increase in sense of belonging was measured at 0.11 standard deviations, occurring at the within-person level.
Twitter use was associated with a marginal increase in affective polarization at the within-person level, suggesting that Twitter usage could slightly exacerbate users’ feelings towards opposing political groups. The effect size was relatively small, indicating that the platform’s role in driving political divides may be more complex than previously assumed. More pronounced was the relationship between Twitter use and outrage, with users reporting a 0.19 standard deviation increase in such emotions.
“The key takeaway from our study is the potential link between Twitter usage and certain adverse effects,” de Mello told PsyPost. “While we cannot definitively assert causality, our study suggests that Twitter usage correlates with diminished well-being, increased outrage, polarization, and sense of belonging. The observation regarding sense of belonging is particularly intriguing, as it indicates that Twitter can also offer benefits, highlighting a complex relationship between social media use and psychological outcomes.”
The researchers also investigated how different Twitter behaviors and functions relate to these psychological effects. Passive uses of Twitter, such as scrolling through the feed, were linked to decreases in well-being, aligning with theories that passive consumption of social media content can be detrimental. In contrast, active engagement, such as replying to tweets or checking trending topics, was associated with an increased sense of belonging.
These findings underscore the importance of how users engage with Twitter, suggesting that active, community-oriented use may mitigate some of the negative emotional impacts associated with the platform.
“Another key takeaway is that how people use Twitter really matters,” de Mello said. “By breaking down the different types of uses, we learned that different usage patterns are linked to different outcomes for the users.”
But the study’s design does not allow for definitive conclusions about cause-and-effect relationships. “Despite the control variables we used, it is still a correlational study, so it’s hard to draw strong causal conclusions,” de Mello noted.
The researchers also noted the potential for selection bias, as their sample included only active Twitter users, possibly excluding those who might have left the platform due to negative experiences. In addition, the focus on short-term effects also leaves open questions about long-term impacts.
“We have a rich dataset of participants public Twitter data (their tweets, timelines, likes, etc.) In our next study, we will analyze the influence of timeline content on self-reported psychological states,” de Mello said.
The study, “Twitter (X) use predicts substantial changes in well-being, polarization, sense of belonging, and outrage,” was authored by Victoria Oldemburgo de Mello, Felix Cheung, and Michael Inzlicht.
