|
Media are once again the worst source of scientific facts. The news teaches us today that playing too much video games "psychologically harms" your children (Too Much Screen Time Harms Kids Psychologically, Too Much TV Time Psychologically Harms Kids: Study). Most of the articles that cite this "fact" do not refer to their sources at all.
After a bit of searching I found the research paper that this news is based on. It is published in the November issue of Pediatrics by a researcher called Angie Page and her fellows.
If you want to check it out yourself, search for "Children's Screen Viewing is Related to Psychological Difficulties Irrespective of Physical Activity". I cannot link directly to the article, as you need permission to access it.
Now, the title of the paper might have tipped you off on why the media drew the wrong conclusion. The researchers found a relation between screen time and psychological difficulties. I think there is also a relation between gravity and us not floating into space.
Do you think that the fact we are not floating into space is causing gravity? It is an absurd example, but it makes clear that you cannot draw cause-effect links between factors when you only know they are related. Also, there could always be a third factor at play that is causing both of the results you see. The researchers mention this weak point in their research themselves:
"The main limitation of this study is its cross-sectional design, because we could not determine whether higher levels of screen entertainment were associated with the development of negative well-being or negative well- being preceded higher levels of screen entertainment."
Also, nearly 2000 kids were selected for the research but only a bit over 1000 were eventually included. For instance, some kids did not fill out the questionnaires properly, and other kids did not follow instructions. Such a huge drop in the sample size makes is hard to believe it was a reliable sample that made it into the research results in the end.
This in turn makes the research less reliable. On top of that, the kids were only tested for 7 days. That was the ideal, but kids with as little as 3 days of testing were also included in the results.
Now consider the real life scenario here: You had a bad week, maybe a fight with a friend, so you don't feel good and do not want play outside. You want to play a game to take your mind off of things. If you are this kid, then you will labeled as having high screen time and low psychological well-being, but is your situation really that unnatural?
Now, I've mentioned some weak points of the research, but it is pretty decent work. Doing psychological research on children is notoriously hard and the researchers did most everything by the book. The only thing that is questionable about this whole affair is the conclusions that are being drawn.
This research shows an interesting correlation between high screen time and low psychological well-being among kids when observed over a week's time. There are loads of possibilities for interesting follow-up research:
Are these results the same when you observe the kids over longer periods of time or are these phases that they go through? Does screen time make kids psychologically worse off, or do unlucky kids get glued to the screen? Does it matter what kids are watching during their screen time?
All in all, the media are drawing completely false conclusions from a decent piece of research. The only thing this research found was that there is some sort of relation between low psychological well-being and high screen time when observing young kids for a few days.
If you want to draw the right conclusions from this, then simply try to raise a happy and balanced child. "Hiding the TV remote and games console controller" only feeds into unfounded hysteria. Common sense goes a long way.
[Cross-Posted from my blog at Think Feel Play.]
|
Very good point :)
What we shouldn't be doing a lot of is feeling guilty about our parenting choices. We did the best we could with what information was available at the time. I agree with you that the early correlations are a little flimsy to get worked up over anyways; however, if the science does become harder and irrefutable over time, I won't be the one arguing against it from an emotional perspective.
Parents have been screwing up their kids for millenia. It's all just one more trip around the sun :)
And my point is not to refute all research that might shed a negative light on video games. There might be follow-up research on this that actually does show that X amount of screen time hurts a child's well-being. To be honest, the line has to be somewhere. There is a limit to everything. Even drinking too much water can kill. What bugs me is how the media feeds into this hysteria by completely misreporting on such research. I saw this reports first on a news source that I rely on. If they misreport this, what else are they getting wrong?
If you have a particular agenda then somewhere there'll be at least one statistic you can wave around as 'proof'. Or if you can't find one, and you have enough money, you can always set up a bogus research facility and make the results you want.
And let's face it, what the media wants is money. A nice, controversial story like this increases their profits. Everyone who clicked on the links to read those articles boosted those news sites' statistics, which feed into their advertising revenue.
FYI...
Like with research on the relationship between watching violence on television and violent behaviour, there are severe ethical limitations that almost completely prohibit any kind of experimental research which could establish a causal relationship here.
i.e. if you start with the premise that an activity (eg. playing video games) could be harmful to children, you can't randomly assign children to different groups and expose them to different levels of that potentially harmful activity.
So correlation data is what we're stuck with. A longitudinal study would give a clearer picture of the causal nature of the relationship, but they're much, much, much more expensive and difficult to conduct.
The large drop in the sample size isn't abnormal, especially when dealing with children and the researchers probably factored that in when designing the study. 1000 participants for a correlational study is reasonably big.
Though, in this particular case, would it really be unethical to let a group of kids play a lot of games for a week, while a control group get to read a book? That way you could look at the short-term effects, while I highly doubt you will change a child's overall development.
With video games and violence I completely agree. I wrote an article exactly on that point on my blog here. Feel free to check it out. It's a rather bold article, and in hind sight I should have more clearly distinguished between "aggression" and "violence". As it is, the article focusses only on actual violence and so the point defending video games is easily made. I'm playing with the idea of writing a separate, more nuanced piece in relation to "aggression".
And about the drop off of participants, 1000 participants is indeed a very good sample. What I was questioning was the quality of that sample when basically unconscientious children are dropped because they fail to fill out the questionnaires and follow instructions. I understand it's a very difficult problem with any type of research involving kids, but I'm not completely sure it cannot be avoided.
Either way, thanks for the informative comment!
I think you're right and it's highly unlikely any long-term harm would come to children so exposed but that's not something an ethics board can be certain of - especially since the premise of the experiment is that it's unknown.
While admittedly much more expensive, a well-designed longitudinal study could answer the same questions as an experiment and do it without the ethical problems. An experiment also can't really address the question of long-term effects and this is the crux.
Re sample quality - the only two issues are randomness (and by extension representativeness of the broader population) and confounding variables. Eg. if the group of children with more screen time had more screen time because they came from families with a lower socio-economic status then other factors associated with lower socio-economic status could account for their results.
I think we all agree that it's good for children to read books. But a kid who spends every moment of spare time reading, never going outdoors or playing with friends, has problems. The same applies to the opposite, a kid who spends all his time playing outdoors and never reading a book. Children have been ruined for life by neglect, and also by overprotective hovering.
Does anyone else remember the brief storm of controversy surrounding the novel Watership Down shortly after it came out? A teenager became obsessed with the book and ran away from home to live in the wild as a rabbit; my memory is shaky as to the result, but he either died or suffered badly from exposure. It's pretty obvious that this particular kid was going to latch onto some obsession sooner rather than later, and Watership Down just happened to be the trigger. But that didn't stop some parents and other do-gooders campaigning for this "dangerous book" to be banned for the good of the nation.
And funny you mention the book reading versus going outside balance. As a teenager I was a very avid reader. I still am, but not to the same degree. My parents encouraged this, seeing reading as something that is "good for you". Then I stayed over with some other part of the family for a few weeks, and suddenly it was a huge problem. They thought it was unhealthy and I should get out more. The positive was turned into a negative, and it became only too clear to me how relative these things are and how skewed some people's views can be. Balance is always key, but a lot of people can't agree where true balance lies.
I'm looking into gaming addiction right now, and I see this question of balance, as well as causation/correlation popping up all over the place.