What's Well & Good in Media

Transcript 27 lines

Cold Open Stats Overview Paper Walkthrough free_promo

Cold Open

Jenny Have you ever closed an app and felt worse, but could not tell which part of it did that to you?

Davis All the time, and my lazy answer is the app did it, but the research keeps pointing at tiny traps inside the feed.

Jenny See, I want to blame the whole app, because that's cleaner than admitting one comment thread or one perfect vacation photo can move my mood.

Davis Right, and if the harm sits in a feature or a social moment, the fix isn't just delete your phone.

Jenny Today we end up with Norwegian teens, where comparing yourself to other people mattered more than just being online across five kinds of mental health symptoms...welcome to What's Well & Good in Media on paperboy.fm.

Stats Overview

Jenny Quick map of the week: the search pulled 1,066 hits, the filter kept 91 papers, and those papers involved 280 authors across 25 countries. So the feed is big, but the actual wellbeing-and-media pile is about ninety studies, not a thousand.

Davis And that gap is the first story. Qualified papers were basically flat, up just 1 from last episode, but total hits jumped by 172, about 19 percent. That says the wider media-and-wellbeing chatter got louder, while the papers that really fit this show barely moved.

Jenny The geography got much narrower, though: 25 countries this week, down from 62, a drop of 37 countries, or almost 60 percent. Indonesia led with 9 papers, then Turkey and China at 5 each, so I'd ask whether this is a real regional cluster or just a search-week artifact.

Davis The topic mix fits the through-line: social media shows up in 24 papers, mental health in 15, adolescents in 8, and then addiction and anxiety at 6 each. That's less, 'is screen time bad,' and more, 'which feature, which age group, which risk, and under what conditions.'

Jenny Methods matter here too. Qualitative studies led with 30 papers, meaning interviews, focus groups, or close reading rather than just counting survey answers. Surveys were next at 23, quantitative work at 16, and cross-sectional snapshots at 10, so a lot of this week can describe experience better than it can prove cause.

Davis The author mix is pretty balanced: 83 first-time authors, meaning their first-ever paper in the metadata, 102 emerging authors, and 95 experienced authors. So the week isn't being driven by one senior network; it's a crowded field trying to pin down where digital life helps, where it harms, and what safeguards actually change the outcome.

Paper Walkthrough

Paper 1 Two Practices That Understate the Mental Health Risks Associated With Adolescent Social-Media Use.

Jenny Alright, let's get into the papers with a measurement fight. J. Haidt and Z. Rausch have a two thousand twenty-six paper in Perspectives on Psychological Science called Two Practices That Understate the Mental Health Risks Associated With Adolescent Social-Media Use, and their basic claim is that the same evidence can look small or serious depending on how researchers slice it.

Jenny Plain version first: weak average links can hide sharper risks for specific kids. They focus on adolescent girls, heavy social-media use, and internalizing symptoms, meaning distress that turns inward, like depression and anxiety. And they say researchers lean too hard on correlation coefficients, which are one-number summaries of how two things move together, instead of elevated risks, which ask how much more likely a group is to cross a meaningful symptom threshold.

Davis So if two researchers look at the same evidence and disagree, what exactly are they averaging together? Are we talking about social media mixed with TV watching, or depression mixed with life satisfaction, or girls and boys put into one bucket?

Jenny Yes, that's the target. They argue that studies often blend exposures, outcomes, and populations in sequence, so social media gets bundled with other digital activities like TV watching, depression gets bundled with broader wellbeing measures, and adolescent girls get averaged with boys. Their point is that this can turn a large, clinically relevant association into something that looks negligible, but the big limitation is that this is a methodological critique of existing literature, not a new dataset with fresh participants.

Davis That makes the takeaway pretty concrete. Before anyone says the effect is small, separate the platform, the level of use, the population, and the outcome. It fits the Beyond Screen Time thread right away, because the argument isn't that every screen is dangerous; it's that a blended average can be too blurry to guide a parent, a school, or a clinician.

Paper 2 Problematic Social Media Use and Depressive and Anxiety Symptoms in Adolescents.

Davis So after that warning about blurry buckets, this next paper gives us a much sharper one. It’s called Problematic Social Media Use and Depressive and Anxiety Symptoms in Adolescents, and it follows eleven thousand two hundred seventy-nine kids in the Adolescent Brain Cognitive Development Study across five annual check-ins.

Davis The plain-language finding is not, kids who use social media are doomed. It’s that teens who had more trouble controlling social media, meaning use that feels compulsive or starts interfering with life, also showed increases in depressive symptoms at specific points over time, especially from year three to year four and year four to year five. Anxiety did not show that same pattern, and the depression links were stronger among female participants.

Jenny How do we know this isn’t just the same kids being more distressed every year, and also rating their social media as more of a problem because everything feels worse?

Davis That’s exactly what the model is trying to separate. They used a random-intercept cross-lagged panel model, which is a mouthful, but it basically compares each teen against their own usual level while also accounting for stable differences between teens. The limitation is important, though: the depression links were year-specific, with no significant lagged link from year two to year three or year five to year six, and they did not find within-person lagged associations for anxiety at all.

Jenny That makes this feel like a strong anchor for the Unequal Youth Risks thread, but not a blank check for panic. Eleven thousand adolescents over five years is hard to wave away, yet the useful screening question isn’t just, how many hours are you online; it’s, is this starting to feel hard to control, and is depression changing around the same developmental window?

Paper 3 Specific aspects of social media use and their associations with symptoms of five mental disorders among adolescents

Jenny That last paper got us to the question of control and depression, not raw hours, and this Norwegian study pushes even harder on that point: Specific aspects of social media use and their associations with symptoms of five mental disorders among adolescents looked at four thousand eleven teens, average age fourteen, across twenty-one lower secondary schools.

Jenny The plain finding is that comparison was the big signal. Teens who used social media in ways that made them compare themselves with others had more symptoms across all five disorder groups, and social comparison explained about twenty to twenty-six percent of the differences in anxiety, social anxiety, and depression symptoms.

Davis So what did they measure besides time spent scrolling?

Jenny They used the short-form Social Media Use Scale, which splits social media into five experiences: getting hostility from others, comparing yourself, passing time, seeking support, and being hostile toward others. Then they used dominance analysis, which is basically a way to ask which predictor carries the most weight when the others are in the model, and comparison came out strongest for boys and girls; getting hostility from others was second at about three to six percent of variance, while passing time and seeking support had negligible independent links. The big caveat is that this is cross-sectional, so it cannot tell us whether comparison causes symptoms, or whether teens with symptoms are more likely to compare.

Davis That feels like the Beyond Screen Time thread in one chart. If you're a parent, school counselor, or platform designer, this doesn't say all social media is poison; it says look for comparison loops and hostility, because those are the features doing more work than just hanging out online.

What's Well & Good in Media

Wellbeing Depends on How We Scroll

Episode