Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
"There Is No Evidence That Social Media Harms Children"
108 Plays1 month ago
Andrew deconstructs the claim that there is no evidence that social media harms children, looking at a widely reported on study from the University of Manchester.
Recommended
Transcript
Intro
Intro: The Dopamine Slot Machine
00:00:11
Andrew Wilmot
Good morning, good day, good evening. Whenever you are, or welcome to the Dopamine Slot Machine, the podcast that discusses what you need to know about the video games your children are playing. How are they designed to get your kids hooked? How do they make money from your children? And what can you do to make sure that your child's relationship with video games is a positive one?
Do Social Media and Mental Health Connect?
00:00:28
Andrew Wilmot
My name is Andrew, I'm a dad of two and a lifelong gamer. And if you're involved in the smartphone-free childhood space, or... actually even if you're not involved in that space at all, you may have heard in the news something to the effect of social media time does not increase teenagers' mental health problems. Research finds no evidence heavier social media use or more gaming increases symptoms of anxiety or depression.
00:00:51
Andrew Wilmot
That is a headline from The Guardian about a study from the University of Manchester released a couple of weeks ago. This is not the only study on this topic, but it does suffer from the same issues that are common amongst these studies, which makes comments such as this one from LSE social psychology professor, government advisor, and obsessive over the digital rights of children, Sonia Livingstone, seem a little bit strange.
00:01:11
Andrew Wilmot
Her statement is, there is no evidence that bans work to make children's lives better, which is why many researchers prefer evidence-based alternatives.
Awareness of Harms by Social Media Companies
00:01:20
Andrew Wilmot
So before I cover the University of Manchester Be Well study, I just want to clarify what no evidence means here.
00:01:28
Andrew Wilmot
It can be easy to think that a lack of evidence found for something means that it is not true, but that's not actually the case. It simply means that no evidence has been found yet. And that can be either because the premise is wrong and that in this case smartphones and social media are actually fine for kids, all because the methodology being used is faulty.
00:01:48
Andrew Wilmot
And I want to further clarify what no evidence means, because we do actually have huge amounts of evidence. Leaked documents from Facebook, TikTok and Snapchat show that this harm has been catalogued, quantified and discussed internally at big social media companies.
00:02:04
Andrew Wilmot
Additionally, in the last few days, TikTok and Snap have settled a high profile lawsuit, which alleges that social media companies intentionally make their products addictive, causing harm to users. By settling now, the evidence to be presented at trial will not come to light.
Insights from the Be Well Study
00:02:18
Andrew Wilmot
It'll be interesting to see if Facebook follows suit.
00:02:22
Andrew Wilmot
Social media companies collect and use mind-boggling amounts of data on their users. This is real data on the real use by real people. If you are on a social media platform, you're an unwitting member of a focus group made up of that entire social network, and your data will be being actively used to refine their platform.
00:02:43
Andrew Wilmot
Secondly, we do have more circumstantial evidence. Jonathan Highton's amazing book, The Anxious Generation, pretty definitively shows 2010 as a threshold of smartphone ubiquity as the starting point of what we call the mental health crisis.
00:02:56
Andrew Wilmot
And we have plenty of evidence to show correlations between social media use and mental disorders. That's not causative evidence yet, at least academically, but we do have, in academic circles, correlations.
00:03:10
Andrew Wilmot
We also know from research on gambling what the impact of slot machine-like mechanics has on adults. Do we need to have a separate study just to confirm that those mechanics are bad for children too? Anyway, so let's look at this University of Manchester B-roll study, which found no evidence that that there was an association between social media or gaming time and mental health.
Limitations of the Study's Scope
00:03:30
Andrew Wilmot
I'm going to quote the article written by the author and published on the site The Conversation, which is a site where academics publish articles are for the more general public. ah It goes, while many previous studies have looked at a single snapshot in time we used a longitudinal approach observing the same young people over an extended period of time we did this through the hashtag bewell programme which surveys young people annually we track the same pupils across three annual waves from year eight when they were aged twelve to thirteen to year nine age thirteen to fourteen to year ten
00:04:04
Andrew Wilmot
aged fourteen to fifteen Another crucial point is that our analysis separated between person effects from within person effects. In other words, rather than just comparing the mental health of heavy users of social media or gaming to that of light users, we looked at whether a specific teenager's mental health worsened after they started spending more time on social media or gaming than they usually did.
00:04:23
Andrew Wilmot
To be clear, I've read the actual study as well and will be referring to text within the actual study, but what I just said was a plain text summary written by the study's actual authors. So firstly, it's a set of surveys and the youngest participants are those in year eight. This is a bit strange to me considering 25% of three to five year olds are active users of TikTok according to Ofcom's latest stats. And we know that there are significant numbers of gamers way below this age. My own children started gaming at about four.
00:04:58
Andrew Wilmot
So to only start looking at the impact of that from 12, unusual. unusual Anyway, let's look at these questions. So firstly for gaming, how ah the the study asks, or rather Be Well survey asks, how often do you play games on a computer or games console such as Nintendo Switch, Xbox or PlayStation? Response options were most days, at least once a week, at least once a month, several times a year, once a year or less, and never or almost never. These responses were recoded into a four-point scale.
00:05:29
Andrew Wilmot
One equals never almost never, once a year less. Two, several times a year, just occasionally. Three, at least once a month. And four, most days slash at least once a week. So immediately from there, two things are noticeable. Firstly...
00:05:44
Andrew Wilmot
the exclusion of gaming on mobile phones or tablets. Younger children in particular are going to be spending more time gaming on those platforms than they will the Switch, Xbox or PlayStation. And we've highlighted on this podcast a number of times the additional risk that comes with gaming on a smartphone or tablet. That being the lack of contextualization for gaming itself. So when you are gaming on a console, you're going to a place that can game.
00:06:08
Andrew Wilmot
And when you're not in that place, you're not able to game. With a smartphone or a tablet, which they could have everywhere, The entire world is contextualized for gaming. Additionally, when you're looking at specifically grooming risks, something like a smartphone or tablet, you have access to other messaging apps.
00:06:27
Andrew Wilmot
And what you often find with somebody who is looking to groom a child is they will seek to move that child away from the game in which they have contact with that child to another messaging platform, typically one which has encryption or disappearing messages. So Telegram, WhatsApp, Signal, might see Discord, Snapchat.
00:06:48
Andrew Wilmot
But the point being, by only focusing on games consoles here, there's a limitation there. Secondly, the highest that a... i child could respond with was that they game most days.
00:07:01
Andrew Wilmot
Now, with the average teenage boy spending three hours a day gaming, it seems strange to put that in with people who game eight hours a day, as you see on the more extreme end, even more than that, with people who might play a bit of chess on their phone on the bus to school. I mean, actually, I've just... On the phone to the bus to school wouldn't be captured by this survey, but The point being, lumping all daily gamers, regardless of time per day gaming, into the same category is... You're going to you're going to miss out on some data there.
00:07:39
Andrew Wilmot
if If somebody goes from gaming 20 minutes a day to gaming 5 hours a day, that is a huge shift that will be missed by this study.
Challenges with Self-Reported Data
00:07:47
Andrew Wilmot
Now for social media. On a normal weekday during term time, how much time do you spend on social media? For example, sites for apps like TikTok, Instagram and Snapchat. Participants selected responses in hourly increments from zero to seven.
00:07:58
Andrew Wilmot
That is better than with the game in question. For those reporting more than zero hours, they had conditional follow-up items asking them to estimate the proportion of time spent engaging in active, such as how much of the time noted above do you spend doing things like chatting with others and posting stories, pictures and videos, versus passive.
00:08:13
Andrew Wilmot
How much the time noted above do you spend doing things like browsing feeds, profiles, or scrolling through photos and stories activities? With the survey set such that responses had to total 100%. It's a good thing that they're trying to split active versus passive in social media. That's going to be better than just looking at purely just social media time.
00:08:34
Andrew Wilmot
But you'll still notice that these questions are asking children how often they play video games and how many hours a day they spend on social media. They're asking them. There's a couple of issues with this. Firstly, it is an observed phenomenon that when survey surveying behaviours where the participants have mixed or negative feelings about the behaviour, even when anonymous, they tend to downplay both the frequency and impact of that behaviour.
00:08:58
Andrew Wilmot
All surveys on social media, gaming or smartphone use are going to fall victim of this because it's a behaviour that is receiving quite a lot of focus at this point in time. It's being widely talked about as a negative behaviour. There is education being done that children are receiving based on the harms and risks of smartphones, social media and games. Now that education is, in my view,
00:09:19
Andrew Wilmot
insufficient, but the point being you're very rare you're not going to find any children who purely positive views on these behaviours. I asked Neil Humphrey, Professor of Education at the University of Manchester, the author of this study, the following.
00:09:35
Andrew Wilmot
Was there any attempt to control for the difference between self-reported use and actual use? He responded to me, re-actual use, no, but worth saying that almost all of the current evidence in this space relies on self-report.
00:09:48
Andrew Wilmot
Let me repeat that again, and this is from the author. That almost all of the current evidence in this space relies on self-report. Again, there is actually one group who is collating evidence based on real data, and that is the people implementing addictive systems.
00:10:07
Andrew Wilmot
So when he's saying that almost all of the current evidence relies on self-report here, he is only correct insofar as evidence put together by
Measuring Mental Health Outcomes
00:10:17
Andrew Wilmot
academics.
00:10:18
Andrew Wilmot
evidence put together by people in industry, the people who are doing this to our children, is based on real data. The quantitative research associate who worked on this study said ah that I was correct that self-reports contained measurement error, but he said that if this misreporting acts as a stable response bias, i.e. it is consistent, then the study accounts for it.
00:10:40
Andrew Wilmot
That is a really big if. and Research on self-reports of self-perceived negative behaviour shows that the participant's reporting accuracy varies according to emotional state, situational factors, and momentary attitudes towards one's device use.
00:10:56
Andrew Wilmot
The existing evidence we've got from adjacent fields show that it is unlikely that this misreporting is consistent, so this is a very serious limitation to this study. There's another less major but still important limitation to the study I wanted to highlight, and that is the way they measure mental health outcomes, which is the me and my feelings measure.
00:11:15
Andrew Wilmot
So they are effectively asking children about their mental health, and that is the degree to which they measure the mental health impacts. Now, we've seen massive increases in teenagers seeking treatment for mental health challenges since social media and smartphones became ubiquitous.
00:11:30
Andrew Wilmot
But this study does not even attempt to measure correlation between social media gaming time and whether an individual seeks mental health support. Now, I also raised this with Professor Neil Humphreys.
00:11:42
Andrew Wilmot
And he said to the following, I'm not sure I agree that treatment for MHDs represents actual mental health outcomes. This assumes that every young person who experiences problems with their mental health actually gets treated, which is definitely not the case. Indeed, the majority don't get treated. So this would have skewed our sample massively. The measure we used, me and my feelings, is very well validated.
00:12:00
Andrew Wilmot
Okay. So firstly, ah The same maths that the quantitative research referred to would have been equally applicable in the, when looking at the skew of data um on people not seeking mental health treatment.
00:12:19
Andrew Wilmot
But it is still a measurable mental health outcome. And if you see that there is a correlation between high gaming and social media use and seeking mental health treatment, isn't that something to be aware of now?
00:12:34
Andrew Wilmot
you know playing devil's advocate against myself here, that might be due to more awareness. And certainly there's a lot of content on social media which is raising awareness of the treatment available for mental health
The Debate on Social Media's Harm to Children
00:12:45
Andrew Wilmot
issues. Whether or not that is a good thing is actually up for debate, but we'll move on.
00:12:54
Andrew Wilmot
Me and my feelings does have similar issues, with the results going to correlate with emotional state, self-awareness and attitudes towards the topic under study, smartphones. So, sorry, social media was the topic under study, as was gaming, not necessarily smartphones.
00:13:08
Andrew Wilmot
So let's go back to the original statement. This study, which is effectively a before and after survey, found no evidence that social media and gaming time has any link with mental health problems. In contrast, the Academy of Medical Royal Colleges has said,
00:13:23
Andrew Wilmot
And quote, without doubt, we are seeing the beginning of a public health emergency with our own eyes. Everywhere we look, we see children and adults glued to their screens. This study has not found any evidence of harm.
00:13:36
Andrew Wilmot
It does not mean that there is no evidence of harm. It is difficult not to read work done by our academics that sanitizes the perception of social media and not get angry.
00:13:48
Andrew Wilmot
Credit to Neil Humphries and his colleagues here. The Be Well study is far from the worst. And there was a further comment he made. He highlighted that a colleague is leading a study where actual data, i.e. scraped from phones, is planned. So watch this space. This is a limitation they're aware of.
00:14:05
Andrew Wilmot
And so credit to Neil Humphries there. I disagree with the applicability of the study that he's led for the reasons given, but he is not one of the worst offenders here. We have a core of digital rights academics who spent years, a decade more or more in some cases, with the focus of their career being justifying children's online time as a positive extension of the child's self.
00:14:29
Andrew Wilmot
What we in the smartphone free childhood movement represent to them is a rejection of their core ideology, that children should be online. Faced with the overwhelming evidence that the foundation of their career, that expansion of children's digital worlds is harming children, it's no wonder that we're getting pushback from them.
00:14:47
Andrew Wilmot
What choice do they have? To admit that they've been active participants in the industrial scale harm to children? To admit their part in what may be the greatest health crisis facing the Western world currently?
Shifting Perspectives on Social Media's Impact
00:14:58
Andrew Wilmot
None of these digital rights academics have ever worked on the development of addictive systems or been a core member of social media product team, but they are as key to the industry as any software engineer. If you had asked me 10 years ago, my stance on the development of social media,
00:15:12
Andrew Wilmot
I'd have argued that it was a good thing with positives for children and that we could build effective controls that allow children to experience the best bits of social media. The evidence of harm, though, now is overwhelming.
00:15:23
Andrew Wilmot
You could be excused for having an optimistic stance on the digital world for children back in 2015. There is no excuse anymore. And so when academics tell us there is no evidence as to the harms of social media, or maybe your MP or head teacher tells you the evidence is mixed, they are wrong.
00:15:43
Andrew Wilmot
They're telling us to ignore the evidence that has been leaked from big tech. They're telling us to ignore the parent on the bus who is too busy scrolling TikTok to have a conversation with their toddler, to ignore the fact that every school in the country is dealing with their children being messaged by strangers on WhatsApp and Snapchat.
00:16:01
Andrew Wilmot
They're telling you to ignore the fact that 13 years after I left school, I can still remember the decapitations, torture videos, illegal pornography and shock pornography that I was shown and sent by peers whilst in school.
00:16:16
Andrew Wilmot
This has been the Dopamine Slot Machine. Thank you and see you soon.