Well I realize all this (have an MS in Stat) but that wasn't my point. The OP was rejecting the study based on his/her own experience in B-school, rejecting the conclusion because it didn't fit with their experience. We'd have to consult the methodology to determine if it was a good study or not, but I don't believe we can reject the results because it doesn't fit with our biased experiences.
So you're disagreeing with results of a survey with 5,000 students across two large countries because you attented one school, with one group of people, and had one group of friends that didn't exhibit the behavior? Is this the type of rigor they taught you in Business school? I'd get my money back.
Why is an academic career the only way to change the world? He should be doing what makes him happiest. Law can be a very academic career, too. Like science, it has a rich history and can be appreicated for more than just a well paying job. Also, he's 18 now? How many people end up doing what they think they want to do at age 18?
Before you get too excited, this is a slashdot link to a zdnet story that links to a blog called Interesting People that posted an email from an end user named David Pollack who got this information from a guy at an 800 number at Windows support. I'll wait until I learn more before making a judgement.
Here's one hack that's missing from the manual. Instead of enclosing a phrase in quotes,/"to be or not to be"/, you can replace the spaces with periods,/to.be.or.not.to.be/. This example turns out to fail, because Google thinks you're looking for a web site in Belgium, but it works most of the time. As a typing-impaired person I like it because it saves having to find the shift key.
Well this is interesting, I think the price concern is still justified considering the systems that beat it were the 3D0 and Neo-geo. Didn't everyone think those systems were ridiculously expsensive at the time, too? It would be more convincing if the PS1 and PS2 prices adjusted to inflation were in line with the PS3 price.
I think your main point is you have to be extremely careful and know what you're doing when using statistics, and I agree 100%! Being a statistician means never having to say you're certain.:)
Well it is highly dependend on what the drug is supposed to do. I work on heart failure trials, so a statement like "Drug A is clinically proven superior to Drug B" could mean something like the length of time for patients on drug A to have a heart attack was significantly longer than that on drug B. You'd have to look at the results of trials for a specific drug if you want to know precisely what was clinically proven. See www.clinicaltrials.gov for a start.
On a related note, what the heck does "clinically proven" mean?
It usually means that a randomized clinical trial was performed and the null hypothesis was rejected. If you haven't had any statistics yet, that means that you hypothesize two treatments have the same effect and run a trial. If the results observed would be highly improbable under the equal effect hypothesis (and this probability is very clearly defined), then you have 'clinically proven' treatment A superior, for instance. Statistical hypothesis testing is basically applying the same principle as a reductio ad absurdum from logic. In general, you usually don't 'prove' things in science like you do in math, but that's another topic.
I have a Master's in Statistics and took a class with Professor Loh about decision trees for multivariate analysis. His kids did go to a public school here in Madison, and one of them won a top ten place in the Intel Science Talent Search competition. On a side note, having him as a father must certainly have helped as he is one of the kindest and smartest people I've ever met.
Meteos is a great game, and if you like Tetris, you'd probably like Meteos. However, as the OP says, you can play Tetris on Nintendo WiFi, against random Internet oponents. You can only play multiplayer in Meteos w/ people in the same room. Internet tetris is really a time killer!
It is worse than that. Your comment is perhaps a bit misguided in two ways. First, even if this poll gets 10,000 particpants, we should still not trust it to provide us with results of the population at large, or even about the population of readers of that site. The reason is that the particpants are not being chosen randomly, they are self-selecting. Therefore, the only thing we can infer from this poll is about the actual participants in it. On the other hand, depending on what effects were interested in and population, 92 participants in a study or poll may be enough to show what we're after. It is entirely dependent on what we're studying and what we are interested in asking and answering. But there is no magic number that works in all situations, such as 30 or 1,000.
After RingTFA, this doesn't seem to be about data mining in the computer science/statistic sense at all. Instead, the article suggests that scientists in academia aren't using the best database tools and techniques available. This I agree with strongly, there is often a disconnect between experiments done in scientific fields and proper database techniques to store that data efficiently. However, I don't call that data mining.
Also, blindly "mining" data for trends can be very misleading. Hypothesis generation is usually better done some other way. There will always be trends in data we already have that are there by chance, and this is what data mining finds in many cases. Then models are fit to that data and don't validate on future samples taken, and everyone wonders why.
Your friend does not start posting at 2, he starts at 1. You and I post at +2 by default because we have the Karma bonus, you can look at that in the moderation history of our posts. He does not, His scores were modded down from 1 to -1, not from 2 to -1.
Where do you get the estimate of 25 points? Previous to the SCO thread, it appears as though only 10 mod points total were spent modding posts down. That's now 2 people we're talking about.
How did she guess a 14 character strong password? Even if it was a word, the chances are extremely remote she'd guess on the first try. I bet there was a keylogger on your system, or some other means besides chance. Too unlikely.
Slashdot Mirror
Well I realize all this (have an MS in Stat) but that wasn't my point. The OP was rejecting the study based on his/her own experience in B-school, rejecting the conclusion because it didn't fit with their experience. We'd have to consult the methodology to determine if it was a good study or not, but I don't believe we can reject the results because it doesn't fit with our biased experiences.
So you're disagreeing with results of a survey with 5,000 students across two large countries because you attented one school, with one group of people, and had one group of friends that didn't exhibit the behavior? Is this the type of rigor they taught you in Business school? I'd get my money back.
Tired of searching for the charger for your rechargeable batteries?
Even with this, VA employees will still be looking for the battery charger.
Why is an academic career the only way to change the world? He should be doing what makes him happiest. Law can be a very academic career, too. Like science, it has a rich history and can be appreicated for more than just a well paying job. Also, he's 18 now? How many people end up doing what they think they want to do at age 18?
Before you get too excited, this is a slashdot link to a zdnet story that links to a blog called Interesting People that posted an email from an end user named David Pollack who got this information from a guy at an 800 number at Windows support. I'll wait until I learn more before making a judgement.
Here's one hack that's missing from the manual. Instead of enclosing a phrase in quotes, /"to be or not to be"/, you can replace the spaces with periods, /to.be.or.not.to.be/. This example turns out to fail, because Google thinks you're looking for a web site in Belgium, but it works most of the time. As a typing-impaired person I like it because it saves having to find the shift key.
Blank stare...
btw, i realize other systems from the 70's and 80's were more expensive, I am talking about the modern era of systems though.
Well this is interesting, I think the price concern is still justified considering the systems that beat it were the 3D0 and Neo-geo. Didn't everyone think those systems were ridiculously expsensive at the time, too? It would be more convincing if the PS1 and PS2 prices adjusted to inflation were in line with the PS3 price.
Those who attempt to gain karma by trying to summarize a complex issue with a one-line quote will have have done neither.
Knowing that their answers may be monitored, what do you think they'd say? These, after all, were done over the phone.
I think your main point is you have to be extremely careful and know what you're doing when using statistics, and I agree 100%! Being a statistician means never having to say you're certain. :)
Well it is highly dependend on what the drug is supposed to do. I work on heart failure trials, so a statement like "Drug A is clinically proven superior to Drug B" could mean something like the length of time for patients on drug A to have a heart attack was significantly longer than that on drug B. You'd have to look at the results of trials for a specific drug if you want to know precisely what was clinically proven. See www.clinicaltrials.gov for a start.
On a related note, what the heck does "clinically proven" mean?
It usually means that a randomized clinical trial was performed and the null hypothesis was rejected. If you haven't had any statistics yet, that means that you hypothesize two treatments have the same effect and run a trial. If the results observed would be highly improbable under the equal effect hypothesis (and this probability is very clearly defined), then you have 'clinically proven' treatment A superior, for instance. Statistical hypothesis testing is basically applying the same principle as a reductio ad absurdum from logic. In general, you usually don't 'prove' things in science like you do in math, but that's another topic.
I have a Master's in Statistics and took a class with Professor Loh about decision trees for multivariate analysis. His kids did go to a public school here in Madison, and one of them won a top ten place in the Intel Science Talent Search competition. On a side note, having him as a father must certainly have helped as he is one of the kindest and smartest people I've ever met.
Meteos is a great game, and if you like Tetris, you'd probably like Meteos. However, as the OP says, you can play Tetris on Nintendo WiFi, against random Internet oponents. You can only play multiplayer in Meteos w/ people in the same room. Internet tetris is really a time killer!
It is worse than that. Your comment is perhaps a bit misguided in two ways. First, even if this poll gets 10,000 particpants, we should still not trust it to provide us with results of the population at large, or even about the population of readers of that site. The reason is that the particpants are not being chosen randomly, they are self-selecting. Therefore, the only thing we can infer from this poll is about the actual participants in it. On the other hand, depending on what effects were interested in and population, 92 participants in a study or poll may be enough to show what we're after. It is entirely dependent on what we're studying and what we are interested in asking and answering. But there is no magic number that works in all situations, such as 30 or 1,000.
I was with you until "McDonald's".
After RingTFA, this doesn't seem to be about data mining in the computer science/statistic sense at all. Instead, the article suggests that scientists in academia aren't using the best database tools and techniques available. This I agree with strongly, there is often a disconnect between experiments done in scientific fields and proper database techniques to store that data efficiently. However, I don't call that data mining.
Also, blindly "mining" data for trends can be very misleading. Hypothesis generation is usually better done some other way. There will always be trends in data we already have that are there by chance, and this is what data mining finds in many cases. Then models are fit to that data and don't validate on future samples taken, and everyone wonders why.
As a counter-example
n etrates_virgin/
http://www.theregister.co.uk/2000/01/10/hacker_pe
Your friend does not start posting at 2, he starts at 1. You and I post at +2 by default because we have the Karma bonus, you can look at that in the moderation history of our posts. He does not, His scores were modded down from 1 to -1, not from 2 to -1.
Where do you get the estimate of 25 points? Previous to the SCO thread, it appears as though only 10 mod points total were spent modding posts down. That's now 2 people we're talking about.
How did she guess a 14 character strong password? Even if it was a word, the chances are extremely remote she'd guess on the first try. I bet there was a keylogger on your system, or some other means besides chance. Too unlikely.
I use R daily. That's usually very annoying in Google. I can search the mailing lists fine, but finding random web pages is usually tough.
Yes, the black bar option is available for games w/out native wide screen resolution. I have used it for BF2 for instance.