There was another attempt to scare people into believing social media is destroying society. That came in the form of a UNESCO study.
We’ve seen this song and dance many times over. Research is done on social media (which is fine), then the results are published (also fine), then any inclination or hint that there is even the most remote possibility that social media is bad gets amplified like crazy by traditional media outlets regardless of the findings (not fine). From the traditional media’s perspective, this is a classic case of a phenomenon of confirmation bias. Sometimes, however, it’s the researchers themselves that suffer from this as well.
In the context of this story, the situation is often that people want to portray social media as this massive nebulous threat to society. Often, it is this fearmongering campaign that dictates that social media is destroying people’s minds and attacking society in general. Therefore, the only solution is to remove social media entirely. This is the starting point for some people out there and it’s just up to the research to catch up to that theory.
There’s, of course, one problem with all of that. As we previously noted, this is not what research is generally finding. Whether it is algorithms radicalizing people, social media making people more anxious, or fuelling depression, research has found time and time again that the data simply don’t back up these stereotypes. This often sparks the mainstream media to try and massage the numbers and jump to a conclusion that simply doesn’t exist because they need ammunition for their war on social media and, by golly, they are going to get it one way or another.
Last year, we saw an incredible example of this. This was found through the US Surgeon General report. Media companies broadcast loudly that social media was fuelling depression among teens. Yet, as TechDirt noted, that wasn’t what the report actually said. It actually painted a very VERY different picture:
Social media can provide benefits for some youth by providing positive community and connection with others who share identities, abilities, and interests. It can provide access to important information and create a space for self-expression. The ability to form and maintain friendships online and develop social connections are among the positive effects of social media use for youth. , These relationships can afford opportunities to have positive interactions with more diverse peer groups than are available to them offline and can provide important social support to youth. The buffering effects against stress that online social support from peers may provide can be especially important for youth who are often marginalized, including racial, ethnic, and sexual and gender minorities. , For example, studies have shown that social media may support the mental health and well-being of lesbian, gay, bisexual, asexual, transgender, queer, intersex and other youths by enabling peer connection, identity development and management, and social support. Seven out of ten adolescent girls of color report encountering positive or identity-affirming content related to race across social media platforms. A majority of adolescents report that social media helps them feel more accepted (58%), like they have people who can support them through tough times (67%), like they have a place to show their creative side (71%), and more connected to what’s going on in their friends’ lives (80%). In addition, research suggests that social media-based and other digitally-based mental health interventions may also be helpful for some children and adolescents by promoting help-seeking behaviors and serving as a gateway to initiating mental health care.
If you want another example, there is also the APA report which TechDirt did a great job at dismantling the lies surrounding that one.
Back in February, research was done on trying to determine whether or not the YouTube algorithm was radicalizing people. The conclusion? No, it wasn’t. When you remove the human element, the algorithm isn’t somehow trying to radicalize people.
So, throughout this, we see a track record of research showing that the scaremongering the mainstream media wants you to believe about social media destroying people’s minds and being an overall threat to society isn’t matching up with the research. This despite efforts to massage the numbers and cherry pick data to try and “confirm” this thinking.
Recently, however, I was noticing yet another study being put on full blast by broadcast media (at least, here in Canada) where the media is trying, yet again, to find this smoking gun that social media is destroying people’s minds and being an overall threat to society. This came into the form of a UNESCO study. The authors, in publishing an announcement about it, made some pretty bold claims about social media – especially TikTok:
Social media negatively affects well-being and reinforces gender stereotypes
Entitled Technology on Her Terms, the report warns that algorithm-driven, image-based content, especially on social media, can expose girls to material ranging from sexual content to videos that glorify unhealthy behaviours or unrealistic body standards. This exposure can have particularly detrimental effects on girls’ self-esteem and body image. In turn, this impacts girls’ mental health and well-being, which are essential for academic success.
The UNESCO report cites Facebook’s own research, which found that 32% of teenage girls said that, when they felt bad about their bodies, Instagram made them feel worse. It also underlines the addictive design of TikTok, characterized by short, engaging videos. This instant-gratification model may affect attention spans and learning habits, making sustained concentration on educational and extracurricular tasks more challenging.
Girls also suffer more cyberbullying than boys. On average, across OECD countries with available data, 12% of 15-year-old girls reported having been cyberbullied, compared to 8% of boys. This situation is compounded by the rise of image-based sexual content, AI-generated deepfakes and ‘self-generated’ sexual imagery circulating on-line and in classrooms. Female students in several countries interviewed for the report said that they were exposed to pictures or videos they did not want to see.
The results demonstrate the importance of greater investment in education – including media and information literacy – and smarter regulation of digital platforms, in line with UNESCO’s Guidelines for the Governance of Digital Platforms, launched in November last year.
What really stuck out to me is the statistic that said that “32%” of teenage girls felt worse about their bodies. My immediate question was, “So, what happened with the remaining 68%?” The excerpt mysteriously doesn’t say. Did the remaining number say that they felt better about their bodies? Was it a split between people who felt neutral and positive? Again, the excerpt doesn’t say. Yet, at the same time, the author felt that this was the number to put right up at the very front so that it takes little time to get that number. That was the red flag for me to start digging deeper.
The report itself, which you can get off of the website or can get here (PDF), gets off to a rather strange start. Specifically, this:
First, it looks at the impact of technology on girls’ education opportunities and outcomes. Although many instances are seen of radio, television and mobile phones providing a learning lifeline for girls, particularly in crisis contexts, gender divides exist globally in both access to technology and in digital skills, although the latter are smaller among youth compared to among adults. Biased social and cultural norms inhibit equitable access to and engagement with technology in and outside of school, with girls always left on the wrong side of the divide.
While technology offers many girls opportunities to access important education content in safe environments, for instance on comprehensive sexuality education, technology in practice often exacerbates negative gender norms or stereotypes. Social media usage impacts learners’ and particularly girls’ well-being and self-esteem. The ease with which cyberbullying can be magnified through the use of online devices in the school environment is a cause of concern, as is the biased design of artificial intelligence algorithms.
For me, these two paragraphs don’t exactly line up. Getting women and girls interested in STEM fields is a good thing. I’m not disputing that. I think people should be free to pursue whatever field they want to pursue regardless of who they are. Yet, the second paragraph seems to jump to the conclusion that social media is stopping women and girls from entering into the STEM field. It suggests that cyberbullying happens, therefore, it’s social media’s fault for women and girls being discouraged from entering into the STEM field. That’s… quite the logical leap, though at the very least, the excerpt doesn’t go so far as to say social media is solely to blame. There’s other factors that happen like people being douche bags in real life, for instance.
Still, any claim that pins social problems on social media should be greeted with skepticism from the get go. This is because it blames technology for human behaviour. If someone sends a nasty text, for instance, do you blame the text service for it or do you blame the person who sent it? Generally speaking, the blame falls on the person sending that text. After all, it wasn’t the texting services fault that the person sent a nasty text in the first place. They can deploy moderation tools and blocking features, sure, but at the end of the day, a person was the one sending that text.
In the early parts of this report, the real question for me is this: “Why is technology, social media, and “algorithms” to blame for gender stereotypes and various other harms?” So far, there appears to be a bit of a disconnect here in showing cause. This next section isn’t exactly something that inspires confidence:
First, our report underlines that men and women have unequal access to information and communication technologies:
130 million fewer women than men own a mobile phone, for instance, and 244 million fewer women have Internet access worldwide – even though digital tools can be a lifeline for girls and women in rural zones, poorer areas, and crisis situations.
Moreover, according to this report, not only are some women and girls unable to access the learning opportunities that the digital transformation may offer, but they are also unable to help shape it on an equal footing.
Indeed, women are currently underrepresented in the technological design and deployment process: in 2022, women held less than 25 per cent of science, engineering and information and communication technology jobs. Today, they represent only 26 per cent of employees in data and artificial intelligence.
This lack of representativity has real consequences on algorithms and data sources, which perpetuate and amplify gender biases. With all too predictable results: according to a recent UNESCO study of generative AI models, a woman is described as a “model” or a “waitress” in 30 per cent of automatically generated texts, while male names are associated with terms such as “business” and “career”.
This situation is also due to pernicious and powerful prejudices among content generators: negative stereotypes paint science, technology, engineering and mathematic as male-oriented fields, causing girls and young women to veer away from STEM career tracks – despite their very real capabilities in these domains.
These stereotypes are also widespread on social media, which girls spend more time on. Girls are therefore more at risk of being exposed to content promoting gendered professions, unrealistic body standards, the sharing of sexually explicit images, cyberbullying – all of which place added strain on their mental health and well-being, and in turn affect academic performance.
All these factors create a feedback loop: where girls are exposed to negative gender norms, steered away from studying STEM subjects, and deprived of the opportunity to shape the very tools that expose them to these stereotypes.
Um, sorry, what? The author of this report highlights a very real issue which is access to technology. In various parts of the world (including Canada), there are very real issues of access. For instance, rural and indigenous communities frequently have a hard time getting access to reliable high speed internet. Thanks in large part to the comfortable ISP oligopolies, those communities continue to be underserved or not served at all. What’s more, it has been demonstrated that if those communities had similar access to the internet as those in more urban settings, it would lead to higher economic productivity as more people turn to entrepreneurship to start their own business and be better able to give back to their communities.
Yet, as you move further down the excerpt, the report basically shifts to more or less arguing that this is all the fault of social media and social media is to blame for these problems (while not making references to anything pointing to these conclusions, I should add). If the author cites access issues that are causing these digital divides, then focus on the access issues. If women and girls are having issues accessing online tools, then how do we address that? That would be an issue I would be on board with.
Yet, to go from “access is a problem” to saying that this leads to cyberbullying on social media is just plain bizarre to me. It doesn’t follow.
Yet, this disconnect is something the author of this report is bent on running with as this was the solutions that were proposed:
The solution, as underlined in our report, starts with education, which plays a major role in re-balancing the gender dimension of technology.
Reducing exposure to social media and negative gender stereotypes. Encouraging more girls to study towards scientific careers through women role models in STEM fields. Ensuring that technological applications are no longer predominantly designed by men. These are some of the recommendations set out in the following pages, which UNESCO is already urging policymakers to implement.
So, the author admits that there are positive aspects to social media, yet concludes that the solution is to reduce exposure to social media in general? That part isn’t making any sense. Now, I’m on board with encouraging women and girls to consider a STEM field. I totally agree that this would very easily help those fields and society as a whole. I think that role models are great. I can happily offer an example in mathematics through MsMunchie123 who is an incredible human calculator. One video example from her is this.
Moving along in the report, there is the repeated evidence that access is a major issue:
Girls and women are on the wrong side of the digital divide.
– Girls and women are less able to access technology. For instance, 81% of men and 75% of women owned a mobile phone in 2023. In Pakistan, 22% of women without a phone cited lack of family approval as the primary obstacle, compared to 4% of men.
– There are 244 million fewer women than men using the Internet worldwide. A survey of girls and their parents in Ethiopia, India, Jordan, Kenya, Nigeria, Rwanda and the United Republic of Tanzania showed that parents tended to believe that girls require more protection than boys from potential online distractions and temptations.
When I look at these points, I’m, again, drawn to the conclusion that there are real access issues. Whether that is through technological hurdles or cultural hurdles, those hurdles do exist. So, when I see that, I think a perfectly reasonable starting point in these conversations is how do we address these hurdles so women and girls are on equal footing to boys and men?
Further, the report also noted the benefits of technology in general:
Technology can help youth access sexuality education that might not be available elsewhere.
– Technology provides a safe and confidential environment for learning about sexuality. The Girl Talk app increased the sexual health knowledge of girls aged 12 to 18. Thailand has e-learning courses for teachers to learn about sexuality education.
– Radio and television have been proven to enhance knowledge on sexual and reproductive health rights. A radio drama in Nepal motivated 11% of those who listened to seek family planning services. Regular listeners of a BBC radio programme in Sierra Leone had a better understanding of the risks of child marriage.
Hey, that’s a good news story. This kind of success should be celebrated, right? Well, none of that really appears to matter to the author as we took a hard turn to arguing that social media is terrible for everyone:
Social media negatively affects well-being and reinforces gender norms.
– Girls are twice as likely as boys to suffer an eating disorder, which is exacerbated by the use of social media. Facebook’s own research found that 32% of teenage girls said that when they felt bad about their bodies, Instagram made them feel worse.
OK, seriously, where is this 32% statistic coming from? What’s the context? This figure was cited one additional time in the report, but only repeats the 32% number. So, we did our own digging and found the Facebook research in question. In that post, Facebook was responding to the Wall Street Journal, pointing out that the outlets conclusions didn’t match the findings of the research. Here’s what Meta stated at the time:
In advance of Facebook’s Global Head of Safety Antigone Davis appearing before a Senate Commerce Subcommittee on Thursday, we want to be clear about what the research recently characterized by The Wall Street Journal shows, and what it does not show.
It is simply not accurate that this research demonstrates Instagram is “toxic” for teen girls. The research actually demonstrated that many teens we heard from feel that using Instagram helps them when they are struggling with the kinds of hard moments and issues teenagers have always faced. In fact, in 11 of 12 areas on the slide referenced by the Journal — including serious areas like loneliness, anxiety, sadness and eating issues — more teenage girls who said they struggled with that issue also said that Instagram made those difficult times better rather than worse. Body image was the only area where teen girls who reported struggling with the issue said Instagram made it worse as compared to the other 11 areas. But here also, the majority of teenage girls who experienced body image issues still reported Instagram either made it better or had no impact. We go into more details below on how the research actually lines up with what The Wall Street Journal claimed.
What’s more, the full chart where the “32%” figure was taken from was this:
Did the report offer the full details of where this number came from? No. Did the report also mention that 45.5% of teen girls said that the platforms had no impact on their body image? No. Did the report mention that these platforms led to 22% of teen girls feeling better about their body? No. Did the report mention that 42% of teen girls felt better about themselves on the subject of social comparison? No. Did the report mention that 51% of teen girls felt less lonely thanks to using social media? No. Did the report mention that nearly 60% of teen girls felt that social media made no difference on the subject of eating disorders? No. What about 57% of teen girls feeling better about their issues with sadness thanks to those platforms? No.
There was literally two whole charts of data to pick from, yet only one single data point was cherry picked from the entire thing. Why not include the rest of the data? There’s very likely a simple explanation: it gets in the way of the conclusion the author wants to push. The author had a conclusion in mind and is working on trying to massage the data to fit with her own preconceived notions rather than letting the data speak for itself. It suggests that there was a considerable amount of confirmation bias going on here – specifically, the author wants to portray social media in the most negative light possible and is willing to cherry pick whatever data point that gives the author the best possible chance of selling this idea.
The section of the report then continues with this:
– TikTok’s algorithm targets teenagers with body image and mental health content every 39 seconds, and with content related to eating disorders every 8 minutes. In the United States, many universities have explicitly banned TikTok from campus.
– Greater interaction on social media at age 10 is associated with worsening socioemotional difficulties with age among girls, while no cross-associations were found among boys.
– Emotional well-being is linked to better academic outcomes. One longitudinal study of families in England found that children with better emotional well-being made more progress in primary school and were more engaged in secondary school.
In this section, there was absolutely no evidence provided to back up these claims. About the only thing I’m aware of that can be corroborated is that there are US universities banned TikTok, but that’s about it.
So, after digging further into the report, we saw some expansion on the TikTok argument:
The addictive design of TikTok, characterized by short, engaging videos, can lead to excessive screen time, distracting students from academic responsibilities and extracurricular activities (Box 8). The platform’s instant gratification model may also affect attention spans and learning habits, making sustained concentration on educational tasks more challenging. For girls, who often face societal pressures to excel academically while also navigating social dynamics online, this can add an additional layer of stress and distraction, making it particularly inappropriate (Amnesty International, 2023; Center for Countering Digital Hate, 2022).
The latter report appears to be related to a report known as Deadly By Design. It doesn’t take long to figure out why that research is especially flawed:
For our study, Center for Countering Digital Hate researchers set up new accounts in the United States, United Kingdom, Canada, and Australia at the minimum age TikTok allows, 13 years old. These accounts paused briefly on videos about body image and mental health, and liked them. What we found was deeply disturbing. Within 2.6 minutes, TikTok recommended suicide content. Within 8 minutes, TikTok served content related to eating disorders. Every 39 seconds, TikTok recommended videos about body image and mental health to teens.
Uh, this is how algorithms, you know, work. You pause, watch, and like content related to eating disorders, then the algorithm will eventually recommend more related content. That is how an algorithm works. That paragraph didn’t say if that content was harmful (it could’ve been perfectly acceptable content about someone talking about eating disorders or mental health), just that it’s the type of content that was recommended.
Yet, that research concluded that it was all harmful:
The results are every parent’s nightmare: young people’s feeds are bombarded with harmful, harrowing content that can have a significant cumulative impact on their understanding of the world around them, and their physical and mental health. TikTok operates through a recommendation algorithm that constructs a personalized endless-scroll ‘For You’ feed, ostensibly based on the likes, follows, watch-time, and interests of a user. CCDH researchers created “standard” and “vulnerable” accounts in the geographies covered.
So, that research could easily be dismissed outright as the researchers in question clearly didn’t understand how algorithms are supposed to work and simply jumped to conclusions.
The previous reference appears to be related to a report titled Driven into Darkness. Apparently, it was based off of an online questionnaire sent to 550 to children an young adults. Not exactly a comprehensive look at how TikTok works. The other report that appears to be the building blocks for the authors report is the I Feel Exposed report. It tries to paint TikTok as some sort of unique surveillance threat, yet the report also admits this:
Amnesty International’s 2019 report, Surveillance Giants highlighted the way in which the abuse of the right to privacy is at the heart of the surveillance-based business model employed by Big Tech companies, particularly social media platforms. This model is based on the massive collection, storage, analysis, and ultimate exploitation of data relating to users who are tracked across the web, through the apps on their phones, and in the physical world, as well through the expansion of the “Internet of Things” as they go about their daily activities. Some apps for instance will track users and collect their approximate or precise location from their mobile device or allow users to “check-in” to locations and share it with their contacts. Other internet-enabled smart home devices such as TVs, doorbells, heating system thermostats etc. may also collect data on users that is aggregated with other personal data.
This has been a long-standing problem with tech companies in general. The obvious solution here is broad federal level privacy reforms to address these issues – something that politicians in North America have been very reluctant to do.
That report also highlights three specific areas of surveillance which are “Explicit targeted advertising”, “Individual personalized targeting mechanisms”, “Lookalike targeting or targeting of lookalike audiences”. These are all very valid points. They are also not unique to TikTok. Meta has long engaged in these activities of targeted advertising for their business model. Google/Alphabet also has systems of targeted advertising and tracking of people through analytics technology. This, once again, highlights the need for broad federal level privacy reform, not that TikTok is some sort of unique threat to our personal privacy.
Of course, all of this is the nuance that the author of the UNESCO report completely glosses over or ignores completely. Again, it is likely because it didn’t quite match up to the theories that the author personally had on this subject. This shows just how much cherry picking is going on here.
The UNESCO went on to offer some conclusions which is just as conflicted as earlier parts of the report:
In principle, technology offers learning opportunities that could overcome certain obstacles, for example in accessing reliable information on comprehensive sexuality education in a safe and confidential environment. But in practice, technology seems to be posing more risks in the opposite direction. Nowhere is this more evident currently than in the way the use of social media affects girls’ well-being and safety, and reinforces gender norms. Algorithms target teenagers far too often with content related to body image and mental health. Greater interaction on social media at age 10 is associated with socioemotional difficulties that worsen with age among girls, while no cross-associations were found among boys. Technology allows for cyberbullying, including through image abuse, which is more common for girls. Ultimately, emotional well-being is linked to better academic outcomes. As is proposed in the recommendations below, education urgently needs to be protected from the negative influences of technology.
The report, at the very least, offers numerous recommendations such as better education and encourage equality in the STEM careers. Again, that’s fine as far as I’m concerned.
The unfortunate aspect in all of this is that many of the reasonable recommendations are going to be completely overshadowed by the highly flawed findings that social media is inherently harmful. In some cases, it relies on other flawed studies that erroneously believe that algorithms are inherently pushing harmful content on everyone or inherently making people depressed, anxious, or worse for everyone when the actual findings showed no such results. Had the anti-social media rantings and bias been removed, this would have been a highly informative report, but unfortunately, personal feelings on the subject matter may have muddied the waters.
The potential effect this could have is that large media companies are going to wave this report around like it was the smoking gun they had long sought to prove that social media must be outright banned from society. Already, Us President, Joe Biden, signed a highly unconstitutional TikTok ban while the Canadian media is openly begging Trudeau to follow suit. These measures are extremely harmful to society for reasons I have already mentioned in those respective reports. As a result, I fear that the mainstream media is going to use this UNESCO report as a launching pad for calling for mass government internet censorship thanks to this flawed research or further justify the censorship measures already being implemented. That will leave everyone behind.
