Violent video game effects on aggression, empathy, ... a meta-analytic review

"Moderator analyses revealed significant research design effects..."

What that means is that the design and sample selection of the studies influenced the outcomes of the studies to a large degree. Oftentimes, we see this happen in studies that draw from clusters of the population. Clusters are not truly random samplings. So when you take a sample population from a cluster -- be it geographic, ethnic, national, political, etc. -- you're run the risk of not taking a genuinely population-representative sample. Your sample population is likely to be more homogenous than a truly random sampling would be, as seen in a lower variance rate among respondents than you'd expect to find from a random sampling.

This is all well and good if what you're testing is the effect of X or Y on that cluster. But if you're trying to study effects of X or Y on the general population, then cluster sampling is far from ideal.

Let's say I'm designing a study called "National Attitudes Toward Ohio State Football." And for my sample population, I select 500 study participants from the state of Michigan. Would anyone in his right mind believe that my study will produce a nationally representative outcome? Similarly, if I'm doing a study called "The Effect of Gas Prices on Grocery Budgets," I might be pre-biasing my outcome if I select from a cluster of long-commute suburbanites. Or from a cluster of poor people. Or a cluster of rich people. Etc.

It's hard to know what sort of design effects were noticed in the various studies aggregated into this particular meta analysis, or how the meta-analysis attempted to compensate for such effects. But the presence of that line in the abstract gives cause for concern.

"Several recent meta-analyses, examining outcome measures most closely related to serious aggressive acts, found little evidence for a relationship between violent video games and aggression or violence. In a new meta-analysis, C. A. Anderson et al. (2010) questioned these findings. However, their analysis has several methodological issues that limit the interpretability of their results. In their analysis, C. A. Anderson et al. included many studies that do not relate well to serious aggression, an apparently biased sample of unpublished studies, and a "best practices" analysis that appears unreliable and does not consider the impact of unstandardized aggression measures on the inflation of effect size estimates. They also focused on bivariate correlations rather than better controlled estimates of effects. Despite a number of methodological flaws that all appear likely to inflate effect size estimates, the final estimate of r = .15 is still indicative of only weak effects."

Meta-analyses are usually either a cheap way to get a well-cited publication or a way to impose one's biases on other people's research. They're rarely as useful as some people would like to think, and the issue a meta-analysis reviews is rarely as simple as some would like to think.

That said, I've got an almost 8 year old, and I obviously don't let him play videos games or watch TV shows with explicit violence (blood, gore, realistic violence, etc.). So for example Mario Kart is okay, but Family Guy is not (they often show bloody violence).

Why is it that whenever someone makes a link between violent video games and violent real life behavior the studies are widely considered to be unfounded, yet folks seem to believe that many of the ills that affect black communities can be directly attributable to aspects of "black culture" (gangsta rap and minstrel media in particular)? There are certainly studies that make the case for the link in both instances, but the conventional wisdom for the two seem to be stark opposites. I've seen these differing beliefs expressed in discussions at Reddit, Ars Technica, and to a lesser extent here.

Why is it that video games can't possibly be a problem, but gangsta rap is most certainly the problem?

(Again, not trying to be a troll, I'm just wondering if there is some fundamental difference that perhaps I've missed. This is your chance to enlighten me.)

Totally surprising that a following meta(!)-study has the same conclusion.

It is known for years that studies in that area are biased (conservative/christian sponsors?) and are often eager to see casualty where none can be proven. A meta-study proves only that such studies exist.