Even highly intelligent people sometimes don't understand technology. It takes a long time to become an expert. Devoting so much time to expertise in something certainly robs one of time to learn other things, like technology.
I new the band before they were famous. Yep...still don't get that mentality at all, but you've aptly called out one of the most annoying geek-culture aspects: measuring one's value on how much more one knows about something over somebody else.
1) I never said you didn't read the article. I merely said it seems like you didn't, based on your circular logic.
2) You've confounded the issue so deeply that I can't separate the article from your posts anymore.
3) Yes, I have issues with the article. More importantly, I have issues with your suggestions about how to improve the article.
4) YOU are the one not addressing what is being said here. You simply feel your logic is sound, all the while you keep throwing out 8th grade level buzz words about design studies. You make it very clear that you really don't know what you are talking about. But hey, keep throwing "double-blind" out there and keep telling ME that I don't understand statistics. I'm married to a statistician, btw. If you don't want to listen to advice, you'll never get anywhere. My research had to go through a month of peer review and revision based on conversations exactly like this one. Sometimes you have to set aside some of your deepest convictions, because in the long run, all they do is invalidate your study with personal bias.
And for the LAST time, this is not about YOU or the article being WRONG--it's about identifying weak factors in a design (be it the article or yours) and bringing to light how these weaknesses invalidate the findings. That's the problem with most people on slashdot...you think everything is either right or wrong, and you argue in a big stupid circle until the end of the Earth.
Now, if you'd like to start over and at least acknowledge that putting a gps in the subjects' cars and measuring the driving speed shows nothing except their driving speed, and that the number of accidents and encounters with police a subject has proves nothing, then we can continue the discussion. Otherwise, I'm tired of hearing you trying to prove that you know something about research design.
You posit three measurements: measuring increase in speed, accident/death rates, and interaction with law enforcement.
I'm simply saying, with total disregard to TFA, that those are three horrible devices, because none of them equate to recklessness on their own and all have other plausible explanations. 1- increased speed could still be well within reasonable speed, speed limits are artificial revenue generators, etc., 2-accidents are not always your fault, and by definition, are not always avoidable, and you can't use "death" because most likely you won't have any deaths, and if you have one death, that won't provide enough data to be useful, or it will skew your results (not to mention the death may be completely accidental and not the fault of the gamer/driver), and 3-cops can pull you over for any non-reckless driving reason--broken tail light, you look suspicious, they can be racist/sexist/ageist.
In all three of these instances you'd have to prove that any of these things (accidents, speeding and interaction with cops) means that a video gamer became a more reckless driver (bringing the conversation back to the context of TFA). Well, first you have to prove that video gaming caused the reckless behavior, but YOU can't do that given you can't even prove that what you consider to qualify somebody as being a reckless driver is actually reckless.
And you can use "Causes" all you want in your study, but that will get your study dismissed faster than BP can make excuses about oil spills.
So, you are saying that if everyone on the planet were to be in the study, that you'd still find that the results wouldn't be statistically significant?
Yes. The design of the study is more important than the number of participants. Nicely summarized. Statistical significance means nothing without context, and in your example, the context is flawed.
The issue of "there aren't enough fatalities" is a statistical problem, not a design one.
No, it's a design problem. Even if you got 200 people to participate (unlikely) the chances of any of them dying during the two years is improbable.
Like I said, all you are gathering is an increase of speed, which on it's own is not an indicator of anything other than an increase of speed.
I don't understand the complaint. Measuring speed accurately and correlating it to gameplay will correlate speed to gameplay.
The article does not say there is a correlation between playing video games and driving speeds. This is why your suggested study does not address the issue in the article. The article makes the claim that people who play video games become more reckless drivers and your proposed study says people who play video games have their driving speed/accident rate/police encounter rate altered to some extent, because of playing the video game.
The article did not say people who play video games become faster drivers. Faster != reckless, unless you define reckless as "driving faster than what is prudent for the driving condition". Using speed limits and encounters with cops is no indication of "recklessness". Your introduction of the irrelevant intervening variables decreases the validity of your design.
You are apparently complaining, not about the study, but about what people who read the results of the study are likely to summarize the results as.
I'm not complaining. I'm offering the insight that your design is inherently flawed because you are using the erroneous conclusion that driving faster makes one a reckless driver. You also overlooked the flaws in the design in that encounters with police and frequency of accidents are also not a reliable indicator of recklessness. If you can come up with a definition of reckless, then measure if drivers really do become more reckless, than you'd have a study. Or, you could change the thesis to something like "playing video games causes people to drive faster". That would be perfectly adequate, and leaves the study open to further discussion and further study--you know, the whole point of research?
Your study is weak if you conclude that a slight increase in speed after playing video games equates to "reckless driving".
And, since I never made that claim, then you must be asserting that my study is flawless.
You didn't? Well damn, what a waste of time then. I could have sworn we were talking about ways to design a study that would support the article's findings of "playing video games leads to reckless driving".
and then measure their average speeds on the way home
And if that's not what you meant, you are saying it now...
Is there any proof that videogames affect driving at all? If not, then this will at least show there to be some effect on average speeds.... And if it shows that there's no link to games and speed, then it's very unlikely that somehow it makes people drive aggressively and unsafely while not driving any faster.
Maybe I'm missing something, but in your previous posts, and in the part above, it is clear that you are making an erroneous assumption that driving faster is equivalent to reckless driving.
I'm beginning to think that you didn't read the TFA, otherwise my points wouldn't seem so cryptic in the context of the article.
I understand statistics just fine. The problem is in the design of the study, not the statistics.
Like I said, all you are gathering is an increase of speed, which on it's own is not an indicator of anything other than an increase of speed. You'd have to prove that an increase in speed has some sort of correlation with bad driving/dangerous driving. Your study is weak if you conclude that a slight increase in speed after playing video games equates to "reckless driving".
Anyway, the whole point of studies is to put it out there for critique. If you don't want to fix your broken design, then fine...nobody will give your study any credence, and nobody will validate it by replicating your findings.
You are mistaken to use average speed as a measurement. That doesn't measure anything other than, well, their average speed. All you'd be able to do is say that drivers who played Game X for X hours drove X faster on the way home. What if they are habitually slow drivers and their new average speed brought them up to the speed limit? What if the speed limits are artificially low and aren't an accurate indicator of dangerous driving?
The largest problem with the Zino is that it is sold by Dell.
You'll go through two or three of them in the lifespan of one Mac Mini, in my experience. Maybe Dell's only suck if you use them at work. I've never been brave enough to try one at home.
I call those "fart cannons". Make your car sound fast, but actually screws up the carefully engineered back pressure, making your car perform even worse.
If the back end of your car loses traction and wants to come around the front end, you can do a couple of things. Steering into the slide AND applying the throttle (for rear-wheel drive) will set the rear end and cause the tires to gain traction, which stops the slide. Force yourself to watch an entire NASCAR event, and you'll see this 100 times per race.
In any case, a lot of things about driving well are counter-intuitive, which explains why there are so many bad drivers. A small example would be the number of people in this thread who talk about how video games taught them to steer with the wheel better. But in real performance driving, you drive more with the throttle input and braking than you do with the steering wheel.
I sort of expect "average" readers to understand correlation. I expect that most people understand 8th grade level concepts like correlation. Maybe my standards are too high?
For me the big thing is getting off the freeway... if I don't stop first, my brain thinks that regular street speeds are way to slow, and I get more of a "damn sunday drivers" feeling right off the freeway.
That's a normal and well documented phenomenon that has nothing to do with racing games. Try the book, Traffic: Why We Drive the Way We Do (and What It Says About Us) by Tom Vanderbilt for reference.
If the ones that played racing games complete the course faster or more recklessly than the ones who played other types of games, then you can demonstrate causation, if not, then you can't.
No, you can demonstration correlation. You said you are normally the one to throw up the correlation != causation flag. If you are, then you should ALWAYS throw up that flag, because it isn't a flag to throw up only when it's convenient to your argument.
...and note the crash rate and fatality rate of your subjects, as well as any encounters with the police for driving activities.
This means nothing. You won't get enough of the control group to die to provide a worthy enough sample, crashes (aka accidents) are sometimes not your fault and unavoidable, and racist cops will screw up the last category. You'd have to subject the groups to a pre/post test model where you can compare the change in their driving ability/behavior.
We could have skipped the entire sensationalized article by skipping to the last sentence: Researchers say the impact of playing games like "Grand Theft Auto" is minimal.
Good example. My friends who are highly religious have a hard time passing the poly, because they aren't lying, but they are nervous about the implications of the questions. I, on the other hand, wouldn't worry one bit if they asked me if I smoked pot (I did 20 years ago). I'd say "no" without a blip, because it's not a big deal.
If I had weekly contacts with Russian agents who paid me lots of money, and they poly administrator asked me "do you have contacts with foreign agents?", then I'd probably have a much harder time passing with a "no" answer--which, my friends, is the ENTIRE PURPOSE of the poly (for the NSA).
MORE importantly, the agency digs up stuff from your past and asks you specific questions about that. "Did you pawn your wedding ring so you could buy drugs?" Were you late on your mortgage for 6 months because of your (drug habit/mistress/gambling problem, etc.).
Seriously, all the conspiracy freaks on here can criticize the polygraph all they want (validly so) in the context of court trials and law enforcement, or even in private employment, but this article is about the NSA's use...yeah, I know, hard to RTFA.
Gee, a Libertarian who distrusts a government agency AND polygraphs? No way!
They use the poly against the subjects as a placebo. The real evaluation comes from the interrogation. It's a good thing the agency is much smarter than you give them credit for, though.
Slashdot Mirror
Even highly intelligent people sometimes don't understand technology. It takes a long time to become an expert. Devoting so much time to expertise in something certainly robs one of time to learn other things, like technology.
I new the band before they were famous. Yep...still don't get that mentality at all, but you've aptly called out one of the most annoying geek-culture aspects: measuring one's value on how much more one knows about something over somebody else.
I have no idea what that means, but you're welcome? :-)
1) I never said you didn't read the article. I merely said it seems like you didn't, based on your circular logic.
2) You've confounded the issue so deeply that I can't separate the article from your posts anymore.
3) Yes, I have issues with the article. More importantly, I have issues with your suggestions about how to improve the article.
4) YOU are the one not addressing what is being said here. You simply feel your logic is sound, all the while you keep throwing out 8th grade level buzz words about design studies. You make it very clear that you really don't know what you are talking about. But hey, keep throwing "double-blind" out there and keep telling ME that I don't understand statistics. I'm married to a statistician, btw. If you don't want to listen to advice, you'll never get anywhere. My research had to go through a month of peer review and revision based on conversations exactly like this one. Sometimes you have to set aside some of your deepest convictions, because in the long run, all they do is invalidate your study with personal bias.
And for the LAST time, this is not about YOU or the article being WRONG--it's about identifying weak factors in a design (be it the article or yours) and bringing to light how these weaknesses invalidate the findings. That's the problem with most people on slashdot...you think everything is either right or wrong, and you argue in a big stupid circle until the end of the Earth.
Now, if you'd like to start over and at least acknowledge that putting a gps in the subjects' cars and measuring the driving speed shows nothing except their driving speed, and that the number of accidents and encounters with police a subject has proves nothing, then we can continue the discussion. Otherwise, I'm tired of hearing you trying to prove that you know something about research design.
Go back to the beginning and start over.
You posit three measurements: measuring increase in speed, accident/death rates, and interaction with law enforcement.
I'm simply saying, with total disregard to TFA, that those are three horrible devices, because none of them equate to recklessness on their own and all have other plausible explanations. 1- increased speed could still be well within reasonable speed, speed limits are artificial revenue generators, etc., 2-accidents are not always your fault, and by definition, are not always avoidable, and you can't use "death" because most likely you won't have any deaths, and if you have one death, that won't provide enough data to be useful, or it will skew your results (not to mention the death may be completely accidental and not the fault of the gamer/driver), and 3-cops can pull you over for any non-reckless driving reason--broken tail light, you look suspicious, they can be racist/sexist/ageist.
In all three of these instances you'd have to prove that any of these things (accidents, speeding and interaction with cops) means that a video gamer became a more reckless driver (bringing the conversation back to the context of TFA). Well, first you have to prove that video gaming caused the reckless behavior, but YOU can't do that given you can't even prove that what you consider to qualify somebody as being a reckless driver is actually reckless.
And you can use "Causes" all you want in your study, but that will get your study dismissed faster than BP can make excuses about oil spills.
So, you are saying that if everyone on the planet were to be in the study, that you'd still find that the results wouldn't be statistically significant?
Yes. The design of the study is more important than the number of participants. Nicely summarized. Statistical significance means nothing without context, and in your example, the context is flawed.
The issue of "there aren't enough fatalities" is a statistical problem, not a design one.
No, it's a design problem. Even if you got 200 people to participate (unlikely) the chances of any of them dying during the two years is improbable.
Like I said, all you are gathering is an increase of speed, which on it's own is not an indicator of anything other than an increase of speed.
I don't understand the complaint. Measuring speed accurately and correlating it to gameplay will correlate speed to gameplay.
The article does not say there is a correlation between playing video games and driving speeds. This is why your suggested study does not address the issue in the article. The article makes the claim that people who play video games become more reckless drivers and your proposed study says people who play video games have their driving speed/accident rate/police encounter rate altered to some extent, because of playing the video game.
The article did not say people who play video games become faster drivers. Faster != reckless, unless you define reckless as "driving faster than what is prudent for the driving condition". Using speed limits and encounters with cops is no indication of "recklessness". Your introduction of the irrelevant intervening variables decreases the validity of your design.
You are apparently complaining, not about the study, but about what people who read the results of the study are likely to summarize the results as.
I'm not complaining. I'm offering the insight that your design is inherently flawed because you are using the erroneous conclusion that driving faster makes one a reckless driver. You also overlooked the flaws in the design in that encounters with police and frequency of accidents are also not a reliable indicator of recklessness. If you can come up with a definition of reckless, then measure if drivers really do become more reckless, than you'd have a study. Or, you could change the thesis to something like "playing video games causes people to drive faster". That would be perfectly adequate, and leaves the study open to further discussion and further study--you know, the whole point of research?
Your study is weak if you conclude that a slight increase in speed after playing video games equates to "reckless driving".
And, since I never made that claim, then you must be asserting that my study is flawless.
You didn't? Well damn, what a waste of time then. I could have sworn we were talking about ways to design a study that would support the article's findings of "playing video games leads to reckless driving".
and then measure their average speeds on the way home
And if that's not what you meant, you are saying it now...
Is there any proof that videogames affect driving at all? If not, then this will at least show there to be some effect on average speeds. ... And if it shows that there's no link to games and speed, then it's very unlikely that somehow it makes people drive aggressively and unsafely while not driving any faster.
Maybe I'm missing something, but in your previous posts, and in the part above, it is clear that you are making an erroneous assumption that driving faster is equivalent to reckless driving.
I'm beginning to think that you didn't read the TFA, otherwise my points wouldn't seem so cryptic
in the context of the article.
I understand statistics just fine. The problem is in the design of the study, not the statistics.
Like I said, all you are gathering is an increase of speed, which on it's own is not an indicator of anything other than an increase of speed. You'd have to prove that an increase in speed has some sort of correlation with bad driving/dangerous driving. Your study is weak if you conclude that a slight increase in speed after playing video games equates to "reckless driving".
Anyway, the whole point of studies is to put it out there for critique. If you don't want to fix your broken design, then fine...nobody will give your study any credence, and nobody will validate it by replicating your findings.
You are mistaken to use average speed as a measurement. That doesn't measure anything other than, well, their average speed. All you'd be able to do is say that drivers who played Game X for X hours drove X faster on the way home. What if they are habitually slow drivers and their new average speed brought them up to the speed limit? What if the speed limits are artificially low and aren't an accurate indicator of dangerous driving?
Except I can't get my PS3 to see my Win7 network, but all my Macs can.
Why in the world are you playing a disk through the computer? The dedicated boxes work pretty well,
So you don't have to have two boxes?
The largest problem with the Zino is that it is sold by Dell.
You'll go through two or three of them in the lifespan of one Mac Mini, in my experience. Maybe Dell's only suck if you use them at work. I've never been brave enough to try one at home.
Or another way to look at it is you get OSX for free when you buy a Mac. Good luck returning it and getting your $0 refund.
I call those "fart cannons". Make your car sound fast, but actually screws up the carefully engineered back pressure, making your car perform even worse.
If the back end of your car loses traction and wants to come around the front end, you can do a couple of things. Steering into the slide AND applying the throttle (for rear-wheel drive) will set the rear end and cause the tires to gain traction, which stops the slide. Force yourself to watch an entire NASCAR event, and you'll see this 100 times per race.
In any case, a lot of things about driving well are counter-intuitive, which explains why there are so many bad drivers. A small example would be the number of people in this thread who talk about how video games taught them to steer with the wheel better. But in real performance driving, you drive more with the throttle input and braking than you do with the steering wheel.
Knowing your car's limit is the best thing anyone can do to become a better driver.
I believe the term you are looking for is "youthful indiscretion". It just takes some hindsight to actually understand what that means.
I sort of expect "average" readers to understand correlation. I expect that most people understand 8th grade level concepts like correlation. Maybe my standards are too high?
For me the big thing is getting off the freeway... if I don't stop first, my brain thinks that regular street speeds are way to slow, and I get more of a "damn sunday drivers" feeling right off the freeway.
That's a normal and well documented phenomenon that has nothing to do with racing games. Try the book, Traffic: Why We Drive the Way We Do (and What It Says About Us) by Tom Vanderbilt for reference.
If the ones that played racing games complete the course faster or more recklessly than the ones who played other types of games, then you can demonstrate causation, if not, then you can't.
No, you can demonstration correlation. You said you are normally the one to throw up the correlation != causation flag. If you are, then you should ALWAYS throw up that flag, because it isn't a flag to throw up only when it's convenient to your argument.
Especially when people use it to dismiss something they disagree with, but otherwise have no productive counter-argument.
...and note the crash rate and fatality rate of your subjects, as well as any encounters with the police for driving activities.
This means nothing. You won't get enough of the control group to die to provide a worthy enough sample, crashes (aka accidents) are sometimes not your fault and unavoidable, and racist cops will screw up the last category. You'd have to subject the groups to a pre/post test model where you can compare the change in their driving ability/behavior.
We could have skipped the entire sensationalized article by skipping to the last sentence: Researchers say the impact of playing games like "Grand Theft Auto" is minimal.
Yes, and I should pour my nasty micro-brew down the sink too and drink a Coors Light. I can't handle the deliciousness. Just like good coffee.
Good example. My friends who are highly religious have a hard time passing the poly, because they aren't lying, but they are nervous about the implications of the questions. I, on the other hand, wouldn't worry one bit if they asked me if I smoked pot (I did 20 years ago). I'd say "no" without a blip, because it's not a big deal.
If I had weekly contacts with Russian agents who paid me lots of money, and they poly administrator asked me "do you have contacts with foreign agents?", then I'd probably have a much harder time passing with a "no" answer--which, my friends, is the ENTIRE PURPOSE of the poly (for the NSA).
MORE importantly, the agency digs up stuff from your past and asks you specific questions about that. "Did you pawn your wedding ring so you could buy drugs?" Were you late on your mortgage for 6 months because of your (drug habit/mistress/gambling problem, etc.).
Seriously, all the conspiracy freaks on here can criticize the polygraph all they want (validly so) in the context of court trials and law enforcement, or even in private employment, but this article is about the NSA's use...yeah, I know, hard to RTFA.
Gee, a Libertarian who distrusts a government agency AND polygraphs? No way!
They use the poly against the subjects as a placebo. The real evaluation comes from the interrogation. It's a good thing the agency is much smarter than you give them credit for, though.