The Diana Award caught up with Amy Orben, DPHIL candidate and college lecturer from the University of Oxford, about her latest research on the effects of social media on young people’s mental health

Q. What do we know about the relationship between social media use and mental health?

“Wearing glasses seems to have the same negative effect as social media use on well-being.”

We talk about the relationship between social media use and mental health a lot. I think that it’s on the forefront of everybody’s minds, but I also think it’s something we know the least about. Both concepts - of social media use and well-being - are incredibly complicated. Social media use is really diverse, it can span skyping your grandmother, to looking at pictures of Kim Kardashian on Instagram to messaging a friend on Facebook. Mental health is really diverse as well, we need to define what makes someone feel ‘good’, how we measure this and at what time we should become worried about decreases in this measurement. We know very little just because of the complexity of what we’re looking at. So, when we’re asking such a general question it almost doesn’t merit finding a clear scientific answer, because there will be such diversity in effects found.

It’s not surprising, then, that research has found both negative effects and positive effects, and effects in the middle and no effect at all.

I think if we would really want to say what we know, it’s that everybody is affected differently by social media – and that’s really important to think about.

When we look at high quality studies, we more often find that technology tends to have a negative effect on mental health. But these negative effects are often tiny.

My colleagues and I try to put this into perspective – by comparing that to other things that might change our well-being. Recent research has found, for example, that eating breakfast and getting enough sleep often times have a lot bigger effect on our well-being, naturally in positive directions, than technology has a negative.

We’re only at the beginning so it’s very important for researchers to be honest about what they don’t know.

Q. Why isn’t our scientific understanding of the effects of social media use better, social media has already existed for quite a long time?

“Bad science is easy, while good science is hard. If you think about good medicine, we know it takes time to get the science right.”

As researchers, we often rely on government, charities and research councils to commission studies and collect data.

Naturally, these bodies start to ask questions when there is a perceived need to, which tends to be a couple of years after the societal change of interest has started.

Collecting data takes time – it also takes time to analyse it properly.

In many ways, when it comes to the tech-mental health issue, we face the normal research challenges, but they’re accentuated by the sense of urgency of understanding technology’s impact on mental health.

Q. Is the lack of shared measurement a problem?

“There’s a huge public debate about mental health and well-being right now. Societies’ standards of what’s accepted change over time, so we don’t actually know: are people becoming more down, or are people becoming more OK to say they’re not OK?”

Measurement is a huge issue.  Often there is a reliance on one-item measures - like asking a person ‘How good do you feel?’ and then taking that as a measure of how good they feel. That doesn’t work. I spent the whole summer working with my colleague who’s a psychometrician, someone who actually studies how people fill out surveys, and I think he had a heart attack every couple of days looking at the way people measure well-being!

If we’re not careful, when asking people questions, especially about sensitive topics like well-being, we can wind up measuring something completely different.

I think it’s also that we have a lot more players in the field now – it’s not just scientists, it’s charities, public intellectuals. We all have access to the data now which means that we need to develop not just a shared understanding of the terms but also a shared understanding of how to use our resources. We have very powerful tools at our disposal now. I sometimes think of having access to this data as akin to having a massive power drill at people’s disposal: how can we ensure that everyone uses it in a way that we can create more knowledge?

Q. What is important to watch out for when evaluating claims made about social media use?

These are the things we need to think about:

  • Effect size: In the medical sciences they’ve already gone towards this a bit more. So, in a medical journal rather than just saying alcohol causes death, you’d have to think about it as: in a general population where everyone drinks a certain amount alcohol, how much more alcohol would people need to drink to increase the death rate? And in a recent paper I try to do just that. I ask: how much more technology would a child have to use to actually feel a decrease in well-being?

  • Source of information: Make sure your information comes from a respectable source. I know a number of UK media outlets do a really good job making sure things are properly reported.

  • The way scientists communicate publicly: Scientist have a responsibility not to exaggerate claims, to contextualise findings, and also to work with press departments and others to make sure that the information that is shared with the public is accessible but also engaging. How do we make really important scientific concepts like “effect size” catchy?

  • The way we seek out information: As readers we also need to work on avoiding just looking for things we agree with.


Q. What should be done to ensure that adolescents are not negatively affected by social media use?

“I think the key is that the solution is built based on young people’s input – and that’s something great that I know The Diana Award does.”

Something that is key for me is communication: that adolescents can communicate about what happens online.

What worries me is that with the current climate of debate, there is so much fear amongst parents, that children may be scared to talk to them openly when negative things happen online. Naturally, there will be things going on which they should have somebody that they can talk to about. However, we concern parents so much that they might try to clamp down on their kids’ tech use, and this can cause adolescents to do it in secret or not feel able to talk openly when something does go wrong because then the parent will say “Oh, I told you so.”

This is a concern of mine, it’s not (yet) backed up by evidence, but it’s something that I think about.

I think the insight we researchers do have is that we can’t ensure that adolescents are not being negatively affected by setting time limits on social media use. I think on the one hand we don’t have evidence of what that time limit should be and on the other hand teenagers are also very resourceful at overcoming barriers to getting what they want.

It’s a really difficult problem. I’m more comfortable saying what shouldn’t be done than what should be done. And that’s key because figuring this out is not a one person job.

This is a hugely collaborative exercise – researchers, people on the ground, practitioners, young people themselves. That’s where government committees where they listen to lots of people are also hugely important.

Q. What are the avenues for future research in this field and how can practitioners like ourselves inform this development?

“Researchers need to work with practitioners, research councils, and others, to push the development.”

For me, what’s key is:

  • There needs to be more talking to each other, including members of the public.

  • As a society, we need to be cautious with numbers. People tend to trust numbers without questioning.

  • As research scientists, we need to not over-generalise from our findings and make sure we’re talking to people on the ground.

  • Different players in the field (research scientists, practitioners, charities, businesses, politicians) each need to maximise what they’re good at instead of trying to do everything. The question though, is what system needs to be in place to make sure that collaboration happens – and that’s a bigger conversation that needs to be had.

Amy Orben’s latest research, ‘The association between adolescent well-being and digital technology use’ can be found on her University of Oxford profile page, here: