In fact, there are a number of rational reasons *not* to make a change. I am a research professor, and time is my field, including chronobiology. I see that someone has already mentioned the heart attacks thing. But there are a number of other effects. Similar to the heart attacks, there is a spike in automobile accidents during time changes. More generally, and probably of more impact, every single person experiences increased stress, and reduced time and quality of sleep due to the change. So it's not only those unlucky enough to be in an accident or have a heart attack that are harmed, it is every single person. (In fact, the increase in heart attacks is almost certainly in large part due to the increased stress, and the increase in accidents is probably due to the sleep disruption).
So, yes, do away with the change. And businesses should not change their operating hours unless they have a direct link to sunrise/sunset. If you're working in an office, there is no need to ever change. If you're working at a park, there probably are good reasons to change.:-)
Yes, exactly. the R squared (variance explained) is tiny. So, yeah, the effect is there, but it's unimportant (and, as you point out, skewed by outliers). There's no assessment of normality of the data (it pretty clearly isn't), which also affects the validity of the results. And, finally, when you have a very large sample size, getting a "significant" result is very easy (20,000 data points is a very large sample size, for statistical purposes). Honestly, with 20,000 data points, I could "prove" pretty much any theory I chose about that data.
Many confounding explanations for the small correlation are ignored that might also have eliminated the observed correlation.
FWIW, I have a PhD, I do this stuff for a living. I got a "significant" result for one of my theories that had an R-squared of 7%. While I of course reported the significance, I also pointed out that it was of no real consequence, and probably due to sample size rather than a real relationship. Especially with the problems of Popper-style hypothesis testing, one should be very careful about what one reports as "real" connections.
I'll second this one. I'm a doctoral student, and have been using it to handle my research. A nice, simple firefox-based interface. It'll snarf references right off pages from search engines. You can attach things, including links to or copies of pdfs to those references, summaries, etc. You can apply keyword tags to citations, and you can organize the citations into a nice directory tree.
To get them out, there's a sweet interface available for open office. I think it's also available for Word, but I use M$ as little as possible.
The only real downsides are copying between computers and citation formats.
Copying is actually easier than it is with the other reference managers I've tried (yeah, I'm talking about you, refworks (bleaargh!) and end note (urrrp!)). You may have to do it more than you would with others, but it's easy to do when you need to. You can export some or all your references to a file, sneakernet the file to the new computer, import it into zotero, and you're done.
There are a lot of citation formats currently available, they just don't happen to be the ones I need. However, there's one close, and the system is designed to be extensible; it's not *that* hard to add your own styles. As soon as I get a round tuit I'll be adding styles for the journals I'll be submitting to, and contribute those back to the project.
Slashdot Mirror
In fact, there are a number of rational reasons *not* to make a change. I am a research professor, and time is my field, including chronobiology. I see that someone has already mentioned the heart attacks thing. But there are a number of other effects. Similar to the heart attacks, there is a spike in automobile accidents during time changes. More generally, and probably of more impact, every single person experiences increased stress, and reduced time and quality of sleep due to the change. So it's not only those unlucky enough to be in an accident or have a heart attack that are harmed, it is every single person. (In fact, the increase in heart attacks is almost certainly in large part due to the increased stress, and the increase in accidents is probably due to the sleep disruption).
So, yes, do away with the change. And businesses should not change their operating hours unless they have a direct link to sunrise/sunset. If you're working in an office, there is no need to ever change. If you're working at a park, there probably are good reasons to change. :-)
Let's make something carried almost universally now by travellers be suspect as a weapon!!!!! The cops would NEVER overreact to that!
"Please step out of your car. I see a weapon mounted on your dashboard."
"What? That's just my cellphone!"
*BLAM*
Yes, exactly. the R squared (variance explained) is tiny. So, yeah, the effect is there, but it's unimportant (and, as you point out, skewed by outliers). There's no assessment of normality of the data (it pretty clearly isn't), which also affects the validity of the results. And, finally, when you have a very large sample size, getting a "significant" result is very easy (20,000 data points is a very large sample size, for statistical purposes). Honestly, with 20,000 data points, I could "prove" pretty much any theory I chose about that data.
Many confounding explanations for the small correlation are ignored that might also have eliminated the observed correlation.
FWIW, I have a PhD, I do this stuff for a living. I got a "significant" result for one of my theories that had an R-squared of 7%. While I of course reported the significance, I also pointed out that it was of no real consequence, and probably due to sample size rather than a real relationship. Especially with the problems of Popper-style hypothesis testing, one should be very careful about what one reports as "real" connections.
I'll second this one. I'm a doctoral student, and have been using it to handle my research. A nice, simple firefox-based interface. It'll snarf references right off pages from search engines. You can attach things, including links to or copies of pdfs to those references, summaries, etc. You can apply keyword tags to citations, and you can organize the citations into a nice directory tree.
To get them out, there's a sweet interface available for open office. I think it's also available for Word, but I use M$ as little as possible.
The only real downsides are copying between computers and citation formats.
Copying is actually easier than it is with the other reference managers I've tried (yeah, I'm talking about you, refworks (bleaargh!) and end note (urrrp!)). You may have to do it more than you would with others, but it's easy to do when you need to. You can export some or all your references to a file, sneakernet the file to the new computer, import it into zotero, and you're done.
There are a lot of citation formats currently available, they just don't happen to be the ones I need. However, there's one close, and the system is designed to be extensible; it's not *that* hard to add your own styles. As soon as I get a round tuit I'll be adding styles for the journals I'll be submitting to, and contribute those back to the project.
Like I said, really sweet, and free.
No, but it'll run NetBSD, of course.