This subject – at this writing, about 3 weeks after I posted Part 1 – almost seems like that of a previous era, with all we’ve experienced since then (see this in The Atlantic from history professor Rebecca Spang). But we, societies around the world, will still be wrestling with this question of humans and screens in the pandemic’s aftermath, so I’m keeping going. Here’s Part 2, taking stock of the challenges researchers face and the corner we seem to have turned….
The New York Times’s Nellie Bowles – who famously reported that the Silicon Valley elite send their kids to small private schools that eschew screens altogether, only confirming so many parents’ concerns – now reports that “coronavirus ended the screen-time debate. Screens won.” She describes a screen-abstinence-only friend who, on Day 3 of sheltering-at-home-with-kids was suddenly “giving them her phone…sitting them at laptops. They were double-devicing. It felt like defeat. Then something surprising happened. They started doing pretty impressive stuff on those screens.”
Now, there’s a turnaround – for a reporter as well as her source (the former even discloses her bias in her earlier reporting) – adults impressed with, maybe even learning from, what children are doing with screens. Will we end up deciding this is an upside or a downside of the pandemic? It’s not either/or, of course, but this quite natural pro-connection response to an isolating force is offering some welcome perspective (scroll to the end for more on that). It took a pandemic. Among many other things, it might be opening minds to what research has been turning up for some time, even as it has been rounding a corner.
One thing the pandemic is making crystal clear is that we can’t really talk about screen use without factoring in the context around it – home and family life, school life – a key argument Dr. Sonia Livingstone made in her transatlantic debate with Dr. Kristi Adamo (Part 1). [Please see the sidebar below for a whole list of research hurdles.] Two other characteristics of today’s media use that I think must make it completely annoying to researchers are that it’s so individual and situational. The meaning of “individual” is obvious, I hope. “Situational” because time is so much a factor in our media use – how and why a person is using (consuming, producing, sharing or interacting with) media and the circumstances around it at a particular moment in time. Because of its spontaneous, very mobile nature now. Capturing and analyzing both of those, individuality and situationality, plus context, across a broad population involves a gargantuan amount of detail. Which is where the “Human Screenome Project” comes in. Probably not the proverbial silver bullet (if there ever will be one), but it’s new, innovative and boldly taking on the individual and situational parts of the research problem.
From genome to screenome
The researchers behind the “screenome” project – Profs. Byron Reeves and Thomas Robinson at Stanford University and Nilam Ram at Pennsylvania State University – wrote in January that telling people to reduce their screen time is like a doctor saying to a patient who’s on multiple medications: “Just reduce the number of pills you take by half.”
They found screen-time-clocking studies so blunt-instrument clumsy that they devised a research methodology “analogous to the genome, microbiome and other ‘omes that define an individual’s unique characteristics and exposures” – in this case, exposures to “apps and websites, the specific content observed and created, all of the words, images and sounds on the screens, and their time of day, duration and sequencing.” They also look at “whether the content is produced by the user or sent from others…and characteristics of use, such as variations in how much one interacts with a screen, how quickly one switches between content, scrolls through screens, and turns the screen on and off.”
[They’re not the only ones challenging “screen time” research, of course. This post would be way too long if it went into how eminent researchers such as Candice Odgers, Andrew Przybylski and Amy Orben have helped the international public discourse turn a corner away from dire generalizations about time in media (see Related Links below for more links and descriptions).]
Two 14-year-old boys’ screens
The screenome project team has so far “collected more than 30 million screenshots – what we call ‘screenomes’ – from more than 600 people,” they report in the journal Nature, of course with plans to collect more.
Exactly what sorts of research insight does that turn up? The instigators give the example of two 14-year-old boys in the same northern California community – individuals whose screen use likely wouldn’t look that different in research relying on self-reports about screen time the day before.
“A typical question that researchers might ask is whether study participants are ‘heavy’ or ‘light’ phone users. Both adolescents might have characterized their phone use as ‘substantial’ had they been asked the usual survey questions. Both might have reported that they used their smartphones ‘every day’ for ‘2 or more hours’ each day, and that looking at their phones was the first thing they did each morning and the last thing they did every night,” the researchers reported in Nature.
But the screenome data turned up significant differences, such as that Teen A had 186 phone sessions each day (a session being the time between when the screen lights up and later goes dark), each lasting an average of 1.19 minutes. Teen B had 26 sessions a day, each averaging 2.54 minutes. Teen A spent 2.6% of his screen time “in production mode” (texting, commenting, posting photos and videos, using a search engine, etc.), while Teen B spent significantly more – 7% – of his screen time “producing.” Boy A used 26 distinct apps, 53.2% social media (“mostly Snapchat and Instagram”), while B used 30 (“mostly YouTube” at 50.9% of his total time. Interestingly, for B, 37% of his screenshots in a single day “included food” – photos from websites, photos of his own food, videos of others eating or cooking and food shown in a game involving running a virtual restaurant.
Even this level of granularity about screen activity seems light on context, though. It may pick up on the possibility that Teen B is a foodie or even potentially a future chef, but – as researcher Mizuko Ito put it in MIT Technology Review’s coverage of the project – it doesn’t say much about how our offline activities shape all that online behavior and vice versa (the Tech Review piece goes into privacy concerns, which the researchers address in their Nature paper).
Contextual, as well as individual & situational
Which brings us back to Part 1. The Human Screenome Project is an important step forward, but it’s not enough. As Ito and Livingstone point out, screen use always has a context. It’s tightly inter-woven with our day, our people, our passions, our life. But the Project is definitely progressive, I think in three ways: It collects and digs into the rich, very individual and situational details of people’s screen use, it offers an important argument from within the research community that I hope intrigues and sparks collaboration with other parts of the community, and it points to greater research insight – especially if somehow combined with context. Screen time, like online safety and wellbeing, is individual, situational and contextual.
As for the pandemic’s effect on our “screen time” sensibilities, the New York Times’s Bowles quotes another parent as suggesting we all – especially our kids – will come out of all this screen-based connecting with renewed appreciation for the human kind. Duh. What I wish more parents saw is that we and especially our children never lost it. Maybe the pandemic will help calm our fears about that.
SIDEBAR: Specific challenges of so much of the screen-time research
Though it had a predictably bias-confirming subhead, an article at Slate.com thoughtfully zoomed in on the challenges that screen-time-clocking research has long faced – issues pointed out by a growing number of influential researchers over the past year. Here’s a partial list:
- Everything but the kitchen sink: If the weather, one’s work, one’s friends, food ordering, the novel you’re reading and your bank – plus 6,000 other activities – are all on your phone, how does a study isolate which one of those might be causing depression, for example?
- Directionally challenged: Say researchers have figured out a way to study just Instagram use, the relationship between it and teen depression. How do they tell if teens are using Instagram because they’re depressed or if they’re depressed because they’re using Instagram? Do a page search in the Slate piece for “hypothesis” and check out that of researcher Jenny Radesky, M.D.
- Causation harder, more expensive to find: “Causal links are difficult to come by; they’re difficult to design and difficult to execute,” Slate reports. So there are a vast number more correlational studies for news stories to cite and policymakers to find, unfortunately.
- The observer’s effect: Now, this is not quantum physics, but human behavior is affected by who’s there observing it. How are research subjects affected by being observed? As Slate put it, “when humans know they’re being studied, that can change how they act.” And what biases and expectations does the observer bring to a study?” [See this about how adults’ fears, not children’s actual experiences, helped set the research agenda in the last decade.]
- About the “clinical” part: If a child’s screen use is being observed in a strange room with lots of pillows, colorful wall art and stuffed animals, etc., it’s not happening how or where it usually happens: at home, in the middle of the family dynamic of the moment. Even if the child’s being observed in the home, the child’s doing their screen thing in front of an adult they hadn’t met before – not how they typically play with or consume media.
- So not medical: “In medical research, it’s possible for both researchers and participants to remain in the dark about who got what treatment, a type of design researchers call a ‘double blind’ study. But with behaviors, it’s impossible to keep things double blind; parents know what “dosage” – how much screen time – their kids are getting.”
- The ethics factor: It’s the question of, if I let my kid participate in a study, how will the study itself affect them? Given that so many “correlational studies [have found] links between heavy device use and negative outcomes,” how ethical would a study be deemed if it required a cohort of children designated for high screen time”? Of course, this is an issue for the Screenome Project too – just good to be aware of.
- Where do screens come in?: The contextual question for screen-time and screenome research: Human sociality is very individual, unpredictable and contextual – whether with screens, with other humans through screens or just in a coffee shop. With young humans, there are so many things influencing that sociality: the social norms and values of their family, their friends, their school environment, other communities their families are involved in, societal and cultural influences, etc., etc. Some of these inputs come through screens, probably most not.
- So, so new: The television version of screen media went mainstream in the 1950s, and we’re still learning about the effects of TV viewing on human beings. And it’s still absolutely relevant – we’re still watching an awful lot of “TV,” just on a bunch of kinds of screens and anywhere/everywhere. Now we need to look at the difference between fixed and mobile TV or big- and small-screen TV. But that’s just media consumption; what’s super new is ever more mobile media sharing, producing and remixing, as well as socializing in the screen media and with it. I don’t see much in the news about how the old ways of using media spill into, overlap and affect the new ways we use it.
There’s so much more to consider. But just on that last point, Slate interviewed Texas Tech researcher Eric Rasmussen about what he learned from studying TV viewing in the home looking at three things: kids’ viewing on their own, viewing with parents and not talking about what they saw, then viewing together and talking about what they saw. The writer doesn’t use the term “experiential learning,” but Dr. Rasmussen’s findings bore out what I learned from UK media professor Stephen Heppell eight years ago. He taught me that co-viewing, playing together or even just talking together about what’s happening on that screen or in that video game can be powerful experiential learning. Parent and child can learn so much about the media, life and each other – our interests, concerns, intelligence, values and dreams – by consuming and engaging with media together.
…to and about research that has led us around the corner:
- In the current moment: MIT Technology Review asks if Covid-19 “has turned back the clock to a kinder time on the web” and, in a commentary in the New York Times, Prof. Andrew Przybylski and psychologist and author Dr. Pete Etchells suggest, “Don’t Freak Out About Quarantine Screen Time.”
- In a review of the research literature, “Adolescent mental health in the digital age: Facts, fears, and future directions” (Journal of Child Psychology & Psychiatry, January 2020), Drs. Candice Odgers and Michaeline Jensen concluded that “most research to date has been correlational, focused on adults versus adolescents, and has generated a mix of often conflicting small positive, negative and null associations. The most recent and rigorous large‐scale preregistered studies report small associations between the amount of daily digital technology usage and adolescents’ well‐being that do not offer a way of distinguishing cause from effect and, as estimated, are unlikely to be of clinical or practical significance. Implications for improving future research and for supporting adolescents’ mental health in the digital age are discussed.”
- Coverage of that study: MIT Technology Review covered the Odgers & Jensen study with “Your kid’s phone probably isn’t making them depressed” and the New York Times with “Panicking About Your Kids’ Phones? New Research Says Don’t.” The Times also cites a forthcoming paper by Jeff Hancock, founder of the Stanford Social Media Lab, presenting his analysis of “about 226 studies on the well-being of phone users,” in which he “concluded that, when you look at all these different kinds of well-being, the net effect size is essentially zero,” the Times reports, later quoting him as saying, “The current dominant discourse around phones and well-being is a lot of hype and a lot of fear. But if you compare the effects of your phone to eating properly or sleeping or smoking, it’s not even close.”
- Here’s the article by Dr. Amy Orben in the journal Social Psychiatry and Psychiatric Epidemiology mentioned above – “Teenagers, screens and social media: a narrative review of [80+] reviews and key studies” – in which she writes, “Reviewing the last decade of reviews in the area, it is evident that the research field needs to refocus on improving transparency, interpreting effect sizes and changing measurement. It also needs to show a greater appreciation for the individual differences that will inherently shape each adolescent’s reaction to digital technologies.”
- The other day: “Why does it suddenly feel like 1999 on the Internet?” asked MIT Technology Review.
- In February: EdSurge senior editor Jeffrey Young interviewed Lisa Guernsey, who literally wrote the book on “screen time” (published in 2012), for “Why Talking About ‘Screen Time’ Is the Wrong Conversation.”
- Over a year ago now, Prof. Sonia Livingstone wrote about the “tipping point” – the pivot away from screen time as a metric in the Parenting for a Digital Future blog of the London School of Economics