Social media negatively impacts teenagers but perhaps only at certain ages 

Written by Ailbhe McKinney, PhD Researcher at the University of Edinburgh and Dr Sinead Rhodes, Senior Research Fellow at the University of Edinburgh.

The association between mental health and social media is complex. It is not as simple as social media does or does not cause poor emotional well-being. There is also emerging research that the relationship goes both ways; that when a person is feeling low, they go on social media more, perhaps to make themselves feel better suggesting it is a coping mechanism for some people. As Ed Sheeran recently said in an interview, “whenever I see a selfie, I want to reach out to that person”. 

A team from the University of Cambridge, University of Oxford, University College London, and Radboud University Medical Center, recently tried to disentangle the complicated relationship between life satisfaction and social media use by examining the different effects it can have on girls and boys and at different ages. 

Key concepts box: Cross-sectional data: data collected at one time. Example: Comparing mental health scores of people of different ages like comparing the mental health of children with teenagers. Longitudinal data: data collected at two or more time points. Example: Comparing mental health scores of the same people at different times. Comparing data collected two years apart from the same people. 

In their study published in Nature Communications, Dr Amy Orben and colleagues looked at cross-sectional data (see key concepts box) with a sample size of 72,287 people aged 10-80 years. They found that the association between social media use (how many hours someone reports they spend on social media a day) and their life satisfaction (measured by a questionnaire) changes with age. Their findings show the strongest negative relationship during adolescence meaning as social media use goes up, life satisfaction goes down. However, with cross-sectional data, you cannot say which came first, just that they are linked as we have described in a ‘how to research the headlines’ post. The same thing affecting someone in a different way at a different age is called a sensitivity period. 

The research team then looked more closely at this sensitivity period with longitudinal data (key concepts box) with 17,409 participants aged 10-21 years. They found a gender difference during adolescence. For girls aged 11-13, an increase in social media use predicted a drop in life satisfaction one year later. The same result was found for boys but a little later at 14-15 years of age. This finding suggests that social media affects girls more at this sensitive period of 11-13 years and affects boys more at 14-15 years. This could be because these are the ages girls and boys go through puberty. Social media use did not predict life dissatisfaction at other ages. 

How did the media report the study? 

This study received a lot of media attention (Forbes, The Guardian, The New York Times, and the BBC ) probably given people’s interest in social media’s impact on young people. 

The methodology was reported well in both The Guardian and the BBC in that they explain one of the main limitations is that the study did not measure what social media young people are using (Instagram vs twitter vs WhatsApp vs TikTok) and what they are doing on them (posting photos, waiting for likes, watching videos of cats, talking to friends). This will be important for the next steps in research on social media and mental health. 

It is good to see the Guardian and the BBC mentioned social media companies’ role in this research in providing accurate data. This point could have been explained a little more by noting that the social media use was measured using self-report assessment i.e. asking how long a person spends on social media opposed to objectively measuring it. 

It would have been good to see the strength of the evidence discussed more across news outlets. The study was strong in the sense that the sample size was large, collected longitudinal data at seven time points and used statistical analysis which checked the direction of the relationship (i.e. which came first). While the study cannot prove that the nature of the relationship between social media and life satisfaction, it provides high-quality evidence. 

How can this study be applied to everyday life?

The study’s findings are based on averages. We cannot say for certain a teenage girl using Instagram at 11 years will become less happy and when she turns 14 years old, she can use social media as she pleases with no repercussions. The study shows what’s known as an effect at the population level meaning not everyone will follow this trend. Social media can be a positive thing in a teenager’s life, especially in a pandemic. Parents should ask their children how it makes them feel and, perhaps pay a little more attention to their child’s use at age 11-13 for girls and 14-15 for boys.