Some points about this metastudy:
i) the discussion of video game violence and agressive behaviour begins not on page 61, but rather on page 56, paragraph 1 - as is to be expected in any published metastudy. This is just a point of order; I'm not claiming that it validates, nor invalidates, the rest of the study.
ii) An interesting note (and the reason for the spoiler tag, as this goes outside the scope of the main thread), is the author's passing reference to what I would deem a very pertinent point, and one which is not explored in any depth: the gender gap.
Playing is heaviest in the 11–14
age group, with boys outplaying girls more than 2.5 hours to 1. A recent
study suggests that around 99 per cent of American boys play video
games, along with 94 per cent of girls (Lenhart et al, 2008). It is common
for US children and adolescents to play more than 20 hours per week
and it is not uncommon for males to play 40 hours or more per week
(Bailey, West & Anderson, 2010).
This 2.5:1 ratio is more important than the authors seem to think it. In fact, this figure informs the rest of the article, without being acknowledged. Consider that the games used as exemplars of antisociality are taken exclusively from traditionally male, or masculine-themed gaming environments (CoD, GTA). I don't think myself particulary guilty of oversimplifying, nor of being disingenuous, when I point out that this particular fact is glossed over as though irrelevant, for the rest of the article. Were it explored, and demonstrated as irrelevant, I would have a much easier time swallowing the rest of the paper. Sadly, it isn't; and this is my first qualm with the methodology - a failure to account for a large statistical difference in gender roles in re: video game playing time, as well as gaming choices doesn't prove a gendered bias in the researchers. It does, however, show a complete disregard for the notion that one must account for one's own biases, and explore/eliminate them as a
precursor to any scientific endeavour - especially one as murky as a metastudy, in as undefined a field as psychology.
Once again - it's not proof that the researchers were wrong; it is, however, evidence that their methodology *may* be biased. I will touch upon this issue again in a bit.
iii) Begging the question. The paper's authors presume the reader is not only aware of, but in agreement with, their own definitions of 'anti-social' behaviour. I will admit to a level of ignorance; there may be a widely-accepted definition within the mental health sciences of which I'm simply not aware.
Before turning to the negative effects of violent video games
however, it is important to stress that video games can have many
helpful benefits. Here are just a few.
'Before turning to the negative effects of (part of A) however, it is important to stress that (wider category 'A') can have many helpful benefits.' Not only is the condescending tone apparent (but that could conceivably be my problem misreading, so I'll ignore it as an argument against this point), but also there is a question of relevance: are the authors claiming that these benefits of the general category 'video game' are somehow negated by the addition of the (undefined) adjective 'violent?' If so, by what mechanism? If not, then how is this to be read as relevant information - especially when there is no comparison of the deleterious effects of 'violent video games' to the positive effects of 'video games?' For example: if the authors' conclusions are true, how are they mitigated/bolstered by these 'helpful benefits?' Is later antisocial behaviour somehow enabled by the effects of pain management? Somehow mitigated thereby? By what mechanisms?
As a (admittedly exaggerated for effect) counterexample: I could attempt to argue that being Jewish is antisocial. If I don't tell you what qualifies or does not qualify as antisocial/prosocial, as relates to my point, then you can't argue against it (post-facto definitions are a great way to sidestep the need for reasoning!) - especially if I can convince you that it's not Jewishness that provides the problem, but rather that it's just, say, Talmudic Judaism. Were that my argument, would adding 'before turning to the negative effects of Talmudic Judaism however, it's important to stress that Judaism can have many helpful benefits' in any way bolster my argument? Especially when, after providing a list of demonstrated benefits of Judaism, I never in any way relate them to my main point, assumed from the outset of my argument, to Talmudic Judaism specifically?
Or, more simply. Squares are bad; here is a list of the benefits of quadrilaterals. How does one relate to the other?
Another point of interest: we're treated to a Wiki'd list of sales figures for 3 games franchises as supporting evidence of the prevelance of (assumed) anti-social content (remember, anti-social is NEVER defined, throughout this article). Yet, when we get to the subsection on 'pro-social behaviour' we're not treated to anything like a named game franchise, to compare the content.
Assumptions: 'Anti-social' is self-evident. This one, I'll allow to slide, with the good faith assumption that there is a strict definition of 'anti-social' used in psychological literature with which I'm simply unfamiliar. CoD, GTA, and WoW, specifically, are exemplars of 'antisocial' video games. <-- With this point, I will disagree, conditional upon familiarizing myself with what definition of 'antisocial' we're using. An unexplored relationship exists between the subcategory Aa ('Talmudic' Judaism, 'violent' videogames, squares) and category A (Judaism, videogames, quadrilaterals) - and providing examples of the most popular forms of Aa is a form of evidence, one which isn't required when asserting points about A. In fact, simply recognizing that A exists is somehow enough to prove that Aa fits my presupposed conclusions about it.
I assert that squares are evil: look at how good parallelograms are. Squares are different from parallelograms. Ergo, squares are evil. QED.
I would be more charitable in my analysis if this read like anything more than a bunch of cherry-picked facts. Not to say there's no validity to the argument; rather, there's no actual argument presented - the conclusion is simply assumed, and facts are thrown at it until some stick, which are then presumed to be evidence.
iv) Weaseling. 'Unfortunately, excessive video game
playing, especially of violent video games, has the potential to impact
children in a number of negative ways.'
Whence comes this notion of 'excess?' Where is the line between 'excessive' and 'NOT-excessive?' We're never told. The entire section of 'Video Game Addiction' amounts to an anecdote, an argument from popularity (inclusion in the DSM =/= factual evidence; or did 'hysteria' suddenly stop being a thing, simply because it doesn't appear in this version? Was it somehow more real or verifiable, simply because the DSM-I included it?), and more frequent references to 'excessive' play - which is meaningless unless the difference between 'excessive' and 'NOT-excessive' is defined.
The entire 'Attention Deficits' section is rubbish, plain and simple. We can claim all the stastically-significant links we want between observed ADHD syptoms and (still undefined!!!!!!!one!111) 'pathological' play - if those doing the observing aren't in any real way qualified to make a medical diagnosis, their observations can (and should be) discounted. Even more telling is that the authors are willing to quote the DSM above, and yet ignore the fact that a diagnosis of ADHD, according to the DSM-IV (and presumably the upcoming DSM-V), requires symptoms to have manifested consistently and habitually from before the age of 7 -
at least a full year before the base cutoff ages of '8-18' from which the authors* draw their data!(* one is tempted to use the term 'novelists,' by this point of the article.)
By the way, Anderson et al's 'belief' in an effect is not evidence.
Especially when the actual data says otherwise.v) This is not a metastudy. This is a rehashing of
this. Basically, we have the primary author quoting the work of the primary author, in order to provide evidence for the assumptions of the primary author.
Try this in the IEEE,
Nature, or the JAMA. There's probably a reason this was published in an online journal. But, I suppose, I'm guilty of snark here, so moving on....
I did read the rest of the paper; unfortunately, nothing was added that in any way validates (or even addresses) the problems outlined above. The conclusion naturally follows from the premises, simply because it's one of them!
