Researcher Connects Playing Shooters With Better Aim

[Treblaine]

Uh, the sample size is large enough, how could you get much more than 150 people for a test like this? Even divided into two groups 75 data points is enough to demonstrate a significant trend.

No, you must realise that the number of students used is unimportant basically, because he is testing something else. He is not testing 'do my students show an average that can be compared to population X'. He was testing if gamers with a gun-shaped controller are more accurate than Joe Average.

And even if he was comparing an average, 150 wouldn't be enough because the sample group was extremely biased, not random at all. If you're working with a biased group, not even thousands and thousands of people are a large enough group. (because no amount of students in his class can form an accurate representation of all gamers everywhere on the world)


But an even more damning criticism was in that I noted how he failed to properly set up what he was actually testing: if gamers hit more often.

The number of shots he let them take are the actual test. Are the hits a result of coincidence, or an underlying trend? And like I noticed, for that case, the sample was way too small, and the method of testing extremely flawed.

Could you explain where this bias comes from?

Is it safe to assume it comes from the "screening" stage where they are interviewed for their gaming habits and their prior experience with fire arms? Are you suggesting he put all the people who had pistol experience in the group to be tested for link between playing light-gun games and "real world" accuracy with an airsoft pistol. Or is it something else?

My problem is finding that playing 20 minutes on a toy (Wii video game) meant they were better with another toy (airsoft) is utterly pointless. The relevance should be playing ANY video game involving aiming and shooting (like Call of Duty on a gamepad) with a firearm most often used in murders (S&W revolver seems most common). I don't have access to firearms nor even the several hundred people to test this but I genuinely would like to see this test. I think the result will give no advantage to the CoD gamer. Yet He gives them a light gun to play with then instead of testing with an actual Pistol that is heavy (almost 3-pounds fully loaded) hard to load and complicated to set-to-fire and that has significant recoil, a heavy moving slide and spitting hot brass. He instead tests them with a light plastic airsoft pistol with no significant recoil.

I understand he wants to take reasonable safety precautions, but how about using a Simunition that the military uses? This is real guns loaded with reduced power paintball loads instead of lead bullets in the cartridges, they are very unlikely to be lethal but still have significant recoil and don't particularly interfere with the pistol's functioning. Or if you can't do that, just have the test performed with a real gun with one round loaded at a time by a Range Officer who constantly stands by the test subject to make sure they don't point the gun somewhere dangerous.

As to the "hit more often" do you mean that in the test the subjects were just allowed to take as many shots as they liked with no particular instruction? It could be simply the game players simply inferred from the game they played earlier that they should aim for the head. That's all this test would indicate, the power of suggestion.

However if it was, all that would indicate was a greater tendency to fire more shots. Did his witness plate (I presume it was a paper target with human silhouette) also record missed shots? The more I look at this "study" the more sloppy and contrived it seems.

Does anyone have an actual link to where his study is published? If it is even published AT ALL! I hope he didn't just put it on his website without any kind of review or assessment process otherwise we have all wasted a lot of our time.

 

[Harker067]

Go reread the link i posted or read it. He let them take 16 shots not 6 with a sample size of about 30 people in 5 groups.

Can't see that anywhere, but even if, that's still barely half of the *absolute minimum* required to draw any conclusions. Any outcomes will still have fallen completely under randomness.

Also, with 16, a fraction of either 33 or 99 is still impossible, so the likelyhood of fraud in the form of fictional data is still present.

Remember: the thing he's testing is a possibility of a connection existing between variable 1 (being a gamer with a gun-shaped controller) and variable 2 (accuracy). That is a simple method 1 examination of testing whether a sample holds up to a population, he can't just Z test it and expect to be done with it.

And in addition, like I wrote, the setup was also fatally flawed and interfering variables are at play.

First off read this article http://www.livescience.com/19984-violent-video-games-improves-real-shooting-accuracy.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+Livesciencecom+(LiveScience.com+Science+Headline+Feed)

While not the paper it does go into some of the details of the study. It clearly says each group fired 16 shots.

Ok so 151/5 groups is ~30 people per group with 16 shots per person (not more or less) that's 480 shots per group a decent sample size. The link sites an average of 7/16 headshots in the gun in gun game group. While we don't have the data or the standard deviation that's still ~210 shots.

The escapist story says that 99% is a comparison to several groups who did not use the gun in the gun game " 99 percent more headshots on mannequin targets using a real gun than those who played the other games".

The link doesn't mention all the numbers but says "Participants who played the nonviolent, non-shooting game had the fewest head shots, an average of about 2. Those who played the other games, including those who played the violent shooting game with a standard controller, fell in between those extremes."

So what this means is that the average of those other trials was a little under 3.5 head shots. They are using the combined data sets in these cases its perfectly legitimate and can easily get you figures like 99% and 33%.

While you may not agree with that comparison being made or the research methods I see absolutely no reason why this looks like manipulated data. While I may not like the results its seems probable that the data they are representing really does show this. If you're going to criticize it please do it a bit more responsibly as you come off as tribalistic and reactionary. Such reactions and sloppy arguments don't help us they just make us look bad.

While I agree that the study is small and preliminary (I have some issues with the idea that it improves accuracy) I don't actually see a huge reason why the aiming for the head part need be wrong. I don't see why being primed by playing a game that rewards head shots then going onto say a fireing range a player might not be more likely to shoot for a target's head. My link mentions the shooting is done immediately after playing the game. I'd personally be more interested in a number of follow up studies. Looking at things like moving target, larger sample sizes, how long after playing a game does any kind of preference for head shots last, does watching a racing movie affect ones likelihood of speeding immediately following.

 

[bandman232]

I've been a target shooter for about ten years now. I can honestly say that games haven't done much to help with my accuracy. If anything, I get stressed and become a worse shot.

Also, I want to know just how close the target range was. Anyone can hit a target in the head at ten feet, but ten yards is a different story.

Hey, I leaned how to operate a few of them. I had no idea how to cock a M16 until I played a moder FPS, because I didn't have AR yet.

 

[ScruffyMcBalls]

I didn't read through all the previous comments, so if this has already been said, I apologise. But his research proves nothing important. Just because these people know HOW to shoot does not mean that they WILL shoot, a human being. You can take the greatest marksman on the globe and put him in a room with a gun and another person for a few hours, nothing will likely happen. But replace that marksman with a psychopath who enjoys pulling the wings off of flies, then we'll have a different (and arguably more "interesting") story. It's like claiming that having a gleaming career in demolition means you're going to commit acts of terrorism, it's bullshit, plain and simple. Anyone who uses this research against the video game community shouldn't have a hand in the debate, also plain and simple.

So, does anyone else want this soapbox? I'm done with it now.