子供と教育 #20：スマホと画面

We have a respite, so I thought I’d tackle various thoughts on children, phones and screens. GPT-5.6-Sol drops tomorrow, and the Fable agents are hard at work.

I’ll start with the other screens, then finish with the phones.

Table of Contents

EdTech.

NonEdTech.

Do Not Ban Social Media Outright.

Some Modern Kids Media Is Pretty Great.

Ban Phones In Schools (1).

Your Offer Is Acceptable.

Ban Phones In Schools (2).

Screen Time.

Inappropriate Content.

EdTech

Increasingly, when you pick a school, you are picking EdTech. The school will put your child on a tablet or computer, and expect them to learn that way.

In theory, with sufficient assistance and bespoke design and incentive structures, this is The Way. It sure seems way better than ‘sit and listen to a lecture.’

I am especially excited for Alpha School’s version of this, with its bespoke designs and high level of both expectations and continuous human support.

Alas, most people are getting a much worse version, that is much worse than what you could easily improvise at home. I’m less concerned with ‘EdTech provider is bilking the system with its torment nexus’ and more concerned with ‘children assigned to spend their days in the torment nexus.’

Ryan Moulton: When you chose a school for your kids, you probably didn’t realize you were choosing educational software, but that choice of software might be more consequential for your kid than the choice of school.

Kelsey Piper: A very well-written, very justifiably angry parental reflection on edtech:

Ryan Moulton: When my son was in first grade, he came home from school in tears saying that he hated math. My wife and I are both engineers, so this was the sort of all-hands-on-deck shock that demanded our immediate attention.

Before this my son had loved math. He would demand that we challenge him with math problems to do in his head in the car and over dinner. He loved doing flashcards. He played math games on his tablet unsupervised for hours.

Even now, years later in 4th grade, he has decided he wants to learn calculus, so he insisted I start explaining it to him as best I could in the car, and started working through pre-algebra in Khan Academy on his own. How is it possible that a kid like this had decided he hated math?

His misery was all due to i-Ready, the software product our district had purchased for math work and testing. During that period my kids’ happiness at the end of the school day was entirely determined by how much time their school had made them spend on i-Ready.

If they hadn’t touched i-Ready, they were happy. If they were forced to do it, they were sad. If they had to spend an unusual amount of time on it, they were in tears.

I started asking around to the other kids’ parents, and I heard similar stories from all of them. Their kids described it as torture. Some of them would hide in the bathroom to avoid it. None of the parents felt that their kids were learning anything at all from it.

…

I have no disagreement with i-Ready’s goals. The problem is that the software simply doesn’t work.

i-Ready assumes that the student cannot read, that they must be read to very slowly, that they must listen to the same instructions hundreds of times, and that they cannot ever be allowed to have any control over this.

Kelsey Piper: The article explains why it was particularly bad as employed in this particular school system, but I don’t think iReady is uniquely bad, and a lot of the things that are wrong here are just things about how edtech in general gets used in schools:

Ryan Moulton: “Being bored” in school is now an entirely different experience than it was when I was a kid. Software enables the enforcement of arbitrary rules that no human being would have the heart or foolishness to enforce.

A teacher, faced with a bored student, would not force them to pay rapt attention to an identical lesson 30 times in a row, 5 days a week, for the entirety of the school year. Software can do that easily. A teacher would not demand that all students take an identical amount of time to finish an assignment regardless of how well they’ve mastered the material.

Software can do that easily. A teacher paying attention to a class will adapt to what is working, what is holding their attention, and what is serving their needs. Software is by default thoughtless, and that allows it to be thoughtlessly cruel.

The full article is even worse.

Dissproportionately: My 14yo daughter read this article and then showed me what iReady looked like for her. Even though she’s in the 8th grade it STILL read everything aloud to her at a slow pace, unskippable.

Alec Stapp: This sounds like torture

Everything Price Sufferer: The CIA literally experimented with a form of psychological torture that was repeating a recording of the same sentence for 24 hours.

Karen Vaites: What if I told you that a @usedgov website was practically marketing iReady Math based on a flawed study conducted during the pandemic?

[goes on to explain more and it only gets worse]

The i-Ready system is used by 14 million students, despite being this stupid.

Our system is such that this level of reaction does not result in ‘oh I guess we should either adjust how we use i-Ready or stop making those kids use i-Ready, then.’

Meanwhile Ryan understood the problem, kept raising the problem, and nothing could be done about the problem.

Ryan literally had to move in order to get his child away from this program, but couldn’t do this until after an entire school year.

How are these schools not being burned to the ground? Or at least having all their students flee in horror?

If this is something that can survive indefinitely, how else do we torture our kids?

Would a teacher force a student to sit in class, 5 days a week, for the whole school year, while you gave a lesson the student already knew and got no value from? Surely the teacher would adjust for that, right?

I mean, maybe, but in many cases no, absolutely not. I speak from personal experience. You can absolutely be assigned such a class in school, as I was, and spend an entire year learning almost nothing, but at least it is not literally the identical lesson 30 times in a row 5 days a week and it wasn’t literal torture.

Matthew Yglesias tries to defend i-Ready, saying his son’s school uses it better so it is not that bad, but to me that only says that school is otherwise so torturous that ‘forced to listen to the same explanations in slow motion over and over again all the time while doing almost zero math’ did not rise to the level of a major complaint. That’s worse, you know why that’s worse, right?

I do agree with the broader point he’s trying to make, that edTech can be good if used well and you should look at the incentives of the system. But, well, look at how it is being used and at the incentives of the system. What a system.

NonEdTech

Even if you do your job at home policing screens, you probably send your kid to school.

Bad EdTech is one problem. Outright hours of random YouTube is another.

Marc Porter Magee: I don’t think a lot of people understand how bad it has gotten in elementary school. Parents are fighting to keep their kids away from screens and junk videos only to have their public schools give away the game

AConcernedParent: We banned YouTube at home only to find out my son was spending 5 hours a day watching YouTube shorts at school on his laptop.

Worst of all, they clearly aren’t even paying for premium.

​NYTimes: A few months before her daughter started kindergarten, Claire Benoist saw a Facebook post that stunned her. Another family with an incoming kindergartner was wondering if it was true that children in the Croton-Harmon School District, 40 miles north of New York City, receive iPads when they start school.

Other parents confirmed that during school, kindergartners often used iPads to play games and watch television shows and YouTube videos. School administrators assured Ms. Benoist that iPad time would be limited to 15 minutes a day, she said. But once school started, her daughter suddenly knew jingles from the diaper and car commercials that would play before YouTube videos she saw in the classroom.

Do Not Ban Social Media Outright

What I oppose as a solution to all this is bans on social media before a fixed age, which is usually 16. Social media bans require age verification techniques that are easy to evade, require large invasions of privacy and lead to frequent leaking of data, and this is a violation of free speech and the ability of children to communicate and in some cases has ended up even extending to Substack. The places kids go instead will often be worse rather than better. This decision should at maximum be up to parents.

Australia tried a ban. It isn’t going great, because you need to hit critical mass, and they failed. Thus, everyone thinks that everyone else is circumventing the ban, therefore everyone circumvents the ban, what like it’s hard?

Tyler Cowen: A few days ago I was talking with a very smart fifteen year old in Australia (really). He was of the opinion that it was quite ineffective, though he noted he could no longer access LinkedIn.

I would note there are more stringent measures, requiring more governmental monitoring and control of the internet, that perhaps could have a greater effect.

The ban including sites like LinkedIn makes it much less likely to stick, because there is far more reason to circumvent the ban. Even if you are happy to give up Instagram and TikTok, especially provided other kids also do so, are you willing to sacrifice half of the internet?

There is one other common argument against such bans that I think is quite bad, which is the ‘why is 16 years old different from 15 years and 364 days’ question. I could ask why one extra day sometimes lets you drink, or vote, or consent to sex. You have to draw a line somewhere, and we presumably can agree that the minimum age for social media or those other activities should not be zero.

I also don’t think rhetoric like this does advocates any favors:

Matt Bateman: “Should children be banned from any exposure to the ultra giga turbo Republic of Letters in which the human spirit is enmeshed and by means of which it is evolving” should obviously be answered in the negative.

Nor do I think that correlational studies should convince us to defy our lying eyes.

reason: For boys, staying off social media might be worse than heavy use, according to a new study. For both girls and boys, using social media *moderately* was tied to better well-being than either abstinence or heavy use.

In the full article, Elizabeth Brown is unusually excellent here about explaining why, and I am glad to report the post is generally very good epistemically.

Elizabeth Brown (Reason): The data revealed “a U-shaped association,” where both social media abstinence and heavy social media use were linked to poorer well-being while moderate social media use was linked to better well-being.

“Among girls, well-being was highest with no use in early adolescence (grades 4–6), but from middle adolescence (grades 7–9) onward moderate use was most advantageous, while high use was consistently adverse and had the greatest association with low well-being in grades 8 and 9,” it elaborates.

“Among boys, well-being was similar for nonusers and moderate users in early adolescence (grades 4–6), but from mid-adolescence (grades 7–9) nonuse became increasingly associated with poorer outcomes, exceeding the risk of high use by late adolescence (grades 10–12).”

… It’s possible that well-adjusted young people with healthy home lives and happy social lives are prone to neither spend too much time on social media nor to avoid it entirely.

… In other words, poorer or better well-being may drive the amount of social media use rather than the other way around—or some third factor (such as super-strict parents) may drive both the amount of time spent online and overall levels of well being.

Brown also points to the years the data covered: 2020 to 2022, which were perhaps the worst time in history to try and go offline.

Here’s a rather scary paragraph:

Keep in mind that the “moderate use” category in this study was not conservative. Up to 12.5 hours per week of social media use between the weekday hours of 3 and 6 p.m. was defined as moderate. So this isn’t merely a finding that the smallest smidge of social media is OK.

There are only 15 weekday hours between 3pm and 6pm. So the students who are ‘heavy’ users need an intervention, and we should freak out that 12.5 hours out of 15 should be considered ‘moderate’ use.

This ‘moderate’ use also includes students who use social media any nonzero amount.

Some Modern Kids Media Is Pretty Great

You can also offer YouTube via whitelisting channels. It can be done.

Jason Crawford: I can empathize with this, and I feel sorry for anyone struggling with digital media and kids. But—

In our household, digital media is a net positive. There are many healthy YouTube channels for our 4yo to watch. Numberblocks is great on math concepts, Daniel Tiger has many helpful life lessons, and even Curious George is wholesome fun. For sure, you have to whitelist the channels, and not let the algorithm descend into slop.

Audiobooks are also great, and access to all kinds of music (mostly Disney), and games. Again, whitelisted, with parental controls.

Maybe this changes for older kids. And probably we have an unusually disciplined toddler who does not get unduly addicted to the iPad. And of course we haven’t gone anywhere near social media yet.

But overall, so far, I’m quite happy to be parenting in the age of digital media.

Andrew Rettek: Numberblocks is the best kids show I’ve ever seen!

Tara Ann Thieke: Numberblocks is amazing. There was a review a few months back which rated Bluey, Paw Patrol etc and Numberblocks was the highest achiever by far. My 4 year old started doing addition all day after a few episodes.

There are some pretty great kids shows now. The production quality these days is off the charts, and some of them really are excellent with the content, especially Numberblocks and Bluey. You also can dip into much of the archives of your own childhood, and a lot of it will not age well. The problem is avoiding the kids getting exposed or attached to the stuff that isn’t good and then demanding it or sneaking it.

You also can pick from the best of the past, on the cheap, and on demand.

Ban Phones In Schools (1)

Allcott et al looked in an unpublished study at impacts from lockable pouches for phones, which I would have thought would be the right way to implement a phone ban. They don’t find much impact in the first year.

In the first year after adoption, disciplinary incidents increase and student subjective well-being falls, consistent with short-term disruption. However, effects on well-being become positive in later years and disciplinary effects fade.

For academic achievement, average effects on test scores are consistently close to zero. High schools see modest positive effects, particularly in math, while middle schools see small negative effects.

We find little evidence of effects on school attendance, self-reported classroom attention, or perceived online bullying.​

In other studies I recall effects tending to improve in subsequent years, as everyone adapts to the new equilibrium, and we see that here as well. Well-being turns positive over time, as does discipline, and it stands to reason this will cash out in academics.

To the extent that this does not reduce student well-being, but also does not substantially improve student academics, that is basically saying that attention paid to school does not improve academics, and that time with phones does not improve well-being. So again it sounds like we should ban, for children, both phones and schools.

Christopher Ferguson analyzes the results in detail, noticing that the authors are being naive frequentists about their conclusions and he questions their statistical approach in other ways as well. He thinks they’re trying to put the thumb on the scale to favor bans and in spite of this still came up with nothing, then coauthor Thomas Dee went to the press saying that this nothing supported bans, warning that some people might draw the opposite conclusion.

The pouches at least are shown to reduce actual use, whereas for most attempted bans the real answer comes from this, another study by Henry Saffer:

Henry Saffer: The outcome variables are screentime and measures of psychological wellbeing. Overall, these early results provide no clear evidence that the school ban policy reduced screentime or improved psychological wellbeing.

Read that again. Banning cell phones in schools had no impact on screen time? What?

I realize that time shifting can be a thing. Kids could in theory be getting more screen time outside of school to compensate for it in school. But the more obvious conclusion is that the bans simply aren’t being enforced at all. So of course they don’t work. Which is indeed a reason not to bother with current bans in their current form.

Here is another perspective, and given the need to keep good teachers it matters:

Arnold Kling: Students using phones in school makes teachers feel disrespected. I am quite certain of that, even if the evidence about the educational benefit of phone bans is contested. If teachers’ feelings count, then as long as getting rid of phones at school doesn’t make things worse for students you can increase aggregate utility by banning them.

Your Offer Is Acceptable

Eli Stark-Elster: Major public intellectuals and politicians have responded by arguing that children should rarely, if ever, participate in digital spaces. As a result, many schools in the US now demand that students seal their smartphones in magnetic pouches. A number of countries, including Australia, the United Kingdom and France, are even considering or have already implemented bans on social media accounts for children and teenagers.

Such restrictions, however, are not the tools of liberation we may imagine them to be. In fact, for some children, the internet may be one of the last remaining spaces where they can grow up doing what children everywhere have evolved to do: independently play and explore with their peers.

Tyler Cowen: Here is more from anthropologist Eli Stark-Elster. I would add a point. I do accept the evidence suggesting that limiting or banning cell phones in schools brings marginally better academic results.

Yet the people who advocate such policies never point out that so many schools are just deadly dull and not very intellectually stimulating? Often what is on the phone is in fact more interesting and sometimes more instructive as well, even if the students do worse in terms of the standards set by the school.

Great point. If phone bans in school don’t do much then school doesn’t do much.

Similarly, a study from Brazil measures impact of a 2023 policy banning non-pedagogical uses of phones in schools in Rio, and shows it lowered phone use substantially but only improved test scores by 0.06 standard deviations.

So, Tyler, your offer is acceptable. We should ban phones from school, and also ban school, by which I mean ban mandatory school. If you choose to go, that’s fine too.

Instead, in response to the study in the previous section, Tyler Cowen seems to think this means we should let kids be on their phones all day instead of school, don’t worry about it, but do still force the kids to be in school while they are doing this.

Tyler Cowen: In sum, it is fine to want to run a school that way, but do not expect huge educational gains, if any. The evidence on this is accumulating, but many seem unable to accept the results. In any case it is not worthy of a major moral crusade.

Saying that a child has to physically be present