I have recently discussed this topic with two different friends.
I like Jonathan Haidt. I really liked his Righteous Mind and loved Coddling of the American Mind. Still, I was skeptical of the causation in Coddling regarding social media and cell phones. There is a great conversation between Tyler Cowen and Haidt where Cowen really pushes back against Haidt. The debate gets a bit heated in the best intellectual way. I was curious if the research and studies in the last two years have supported Haidt or Cowen.
How Bad Is Social Media for Kids
About a month ago, a meta-analysis of over 2.8 million children was published. It found:
"Data from 23 meta-analyses, representing 2 817 252 children from 64 countries, were extracted and synthesized. Screen-based media exposure was associated with lower academic achievement (r = −0.15; 95% CI, −0.23 to −0.07) and language skills (r = −0.14; 95% CI, −0.18 to −0.10) and with higher internalizing behavior (general screen use: r = 0.10; 95% CI, 0.07-0.13; social media use: r = 0.09; 95% CI, 0.06-0.12) and externalizing behavior (r = 0.11; 95% CI, 0.08-0.14), emotional problems (r = 0.30; 95% CI, 0.08-0.50), and peer problems (r = 0.28; 95% CI, 0.04-0.49). Screen-based media exposure was not associated with executive functioning (r = 0.01; 95% CI, −0.09 to 0.11), prosociality (r = −0.22; 95% CI, −0.43 to 0.01), or mathematics (r = −0.10; 95% CI, −0.23 to 0.03), science (r = −0.03; 95% CI, −0.15 to 0.11), or reading (r = −0.02; 95% CI, −0.12 to 0.09) scores. Screen type, subtype of behavior problems, publication year, and age moderated some of these associations."
Here is a more visually appealing view of the information above:
The two largest correlations also have the two largest confidence intervals and have the smallest sample sizes. That means that these associations could change quite a bit with further research. The following table comes from a review of the above mentioned meta-analysis- click here for the original).
| Outcome | Studies | Children | r (95% CI) | I² |
|---|---|---|---|---|
| Academic achievement | 27 | 151,989 | −0.15 (−0.23 to −0.07) | 99.0% |
| Math achievement | 17 | 40,157 | −0.10 (−0.23 to 0.03) | 98.4% |
| Science achievement | 6 | 36,208 | −0.03 (−0.15 to 0.11) | 97.9% |
| Reading | 5 | 13,890 | −0.02 (−0.12 to 0.09) | 94.0% |
| Language | 51 | 34,666 | −0.13 (−0.17 to −0.09) | 82.8% |
| Executive functions | 16 | 7,577 | 0.01 (−0.09 to 0.11) | 98.6% |
| Internalizing behavior (general screen time) | 88 | 390,576 | 0.10 (0.07 to 0.13) | 98.2% |
| Internalizing behavior (social media) | 71 | 273,392 | 0.09 (0.06 to 0.12) | 96.1% |
| Externalizing behavior | 84 | 147,976 | 0.11 (0.08 to 0.14) | 95.4% |
| Emotional problems | 9 | 75,413 | 0.30 (0.08 to 0.50) | 99.7% |
| Peer problems | 9 | 74,924 | 0.28 (0.04 to 0.49) | 99.8% |
| Prosocial behavior | 4 | 8,773 | −0.22 (−0.43 to 0.01) | 99.1% |
Overall these findings look concerning. But what do the numbers mean? If anyone is really interested they should read the paper and the review.
I looked for a couple other behaviors to anchor these numbers for people who never took or don't remember statistics.
Behaviors for Comparison
Sleep and physical activity have similar studies and findings. I don't know anyone questioning the effects of sleep or exercise on health, so this should help readers understand the effects better.
The correlations look about the same. It is safe to say, based on the studies produced, that social media and screens have similar correlations as physical activity and sleep. Or rather the opposite, sleep and physical activity are negatively correlated with depressive symptoms while social media and screens are positively correlated.
TLDT
Exercise good. Social media bad.
Causation
The review I shared above determined the studies could not prove causation. Even for the physical activity and sleep studies have the same problem. Proving causation is hard.
It's always possible that depressed people sleep less, workout less, and use social media more.
There are controlled studies with social media. But the controlled studies are usually short term; subjective reporting; and on adults- mostly college students. So it's unclear how long the effect might last and whether they apply to kids (although they very well could apply more!).
Conclusion
With that said, I think there is enough evidence to generally say social media is not good for kids. It is anti sleep and anti exercise.
Context does matter. It maters what they are doing. Who is or isn't supervising and or aware of the usage. Who their peers are. Who their parents are. What time they use it. Where they use. If it was possible to control for a lot of these variables, the picture might be more clear. Something more like, social media is very bad when kids and teens are unsupervised.
Peer groups matter a lot.
This leads me to think that social media in itself is not that bad. I'll grant Haidt the fact that studies continue to support his claims. But he should be more modest with the effect sizes and uncertainty of the causation.
Citations
Al-Jbouri E, Nivison MD, Neville RD, et al. Screen Use and Child and Adolescent Development and Mental Health Outcomes: A Quantitative Umbrella Review. JAMA Pediatr. Published online September 28, 2026. doi:10.1001/jamapediatrics.2026.4339
PCIMH Editorial Team. (2026, September 30). Does screen time harm child development? What a review of 2.8 million children found. Pacific Center for Integrative Mental Health. https://www.pcimh.org/screen-time-child-development-mental-health/
No comments:
Post a Comment