Tweeting throughout the night explained almost 2% of variation in participants’ mental wellbeing, which is comparable to activities like binge drinking and smoking marijuana (as measured in previous studies).
Researchers suggested that actively using Twitter during the night could both disrupt and delay sleep, which could reduce the quality and quantity of sleep, harming mental wellbeing. Nighttime tweeting showed a weaker relationship with depressive and anxiety symptoms (compared to mental wellbeing), although this became stronger after results were split by age and sex.
Seventy-four per cent of UK adults keep their phone in their bedroom at night, while twenty-six per cent say they would check their phone if they wake up in the night, according to a 2022 YouGov survey.
Regulation and guidance for nighttime social media use
The study’s findings support calls for more regulation and guidance for nighttime use of social media. For example, TikTok, the online video-sharing app, introduced the tool “Wind Down” in March this year which shows meditation videos at night to encourage younger users to stop scrolling.
Researchers say top-down approaches to change the user architecture of apps, like TikTok’s wind-down mode, as well as education campaigns to raise awareness within vulnerable groups, could help improve the safety of social media use.
Daniel Joinson, Doctoral Researcher and lead author of the paper said:
“Whilst social media is often treated like a monolith, its impact on mental health will depend on the exact behaviours the user performs and the experiences they have on these platforms. Our paper highlights the potential harm of a very specific behaviour: nighttime content posting.
“Research like ours could help inform interventions or legislation that aim to deter harmful social media use, whilst enabling beneficial behaviours or experiences. This is made possible by having access to actual social media data, which is essential if we are to build a deeper understanding of the relationship between social media and mental health.”
Novel data collection approaches
The research drew on longitudinal data from 310 adults (aged between 18-60+) in the Children of the 90s study who consented and were eligible to share their Twitter data, with 18,288 tweets included in the data. Participants’ mental health was measured at multiple timepoints using standard questionnaires, including the Short Mood and Feelings Questionnaire (SMFQ). Importantly, instead of classifying people as simply depressed or not, mental health was measured on a scale, giving a more detailed picture. Participants’ tweets from within two weeks of these questionnaires were included in the analysis, but all others were not.
Uniquely, the study used data directly from Twitter (with the consent of the participant). This enabled the researchers to collect precise measurements of the time of day participants posted on Twitter.
However, the authors noted that the study participants were all adults, almost entirely white, and were more likely to be female. This data was collected during the COVID-19 pandemic, a unique time for social media usage and mental health patterns.
The research team are now looking to understand more about how the patterns of emotion expression and social interactions relate to mental health and wellbeing.