>The U.S. Surgeon General was recently asked to study the effects of media violence in children. He said they already did. The results are virtually unanimous: of 1000 studies, only 18 found no link between media violence and violence in youth.
I noted in a review of this data that they had failed to factor out things like time spent with parent. E.g. if kids do *anything* (say, watch "I love Lucy") to the exclusion of spending time with role models, they have a greater risk for violence and other anti-social behavior. It's very easy to get a correlation with "violence" which is actually merely a correlation with time in front of the TV & other baby-sitters instead of time with family & friends.
>I don't remember who stated that an X crash leaves Linux in an unusable state, but it's completely false
You're dead wrong. You assume that the X hot keys are still working, that the machine is networked, or that you have a 2.2 kernel with the sysrq keys compiled in & sysrq K actually manages to get X unstuck.
None of these are necessarily true. In particular, it's absurd to say "well, just buy two computers, so when X dies you can telnet in from the other one and fix it!" yeah, right. X shouldn't leave the machine stuck in a state where you *need another machine* to fix it.
I just ran another X hanging program (gnuplot a very large data file). When it locks, the ctrl-alt keys are *not* working. I do have sysrq, and sysrq K appears to have killed X (assuming it didn't die on its own) & since I have xdm running it was respawned.
However the ctrl-alt keys still don't work & in fact lock up X, now. ctrl-alt-f1 freezes the display. and following this sysrq K no longer works.
only sol'n is reboot or log in remotely (since i'm at work & have a dozen machines on the network). killing X via remote login, or going to run level 3 still doesn't get the consoles back. hot keys still don't work.
this is some pretty serious pathology for something as simple as "plot 'data'".
>we don't have laws like many European countries limiting things like how many foriegn movies our theaters can show.
No, in the US we have this cool rating system so tweaked that frequently we can't show uncut american made films in america, though you *can* show the uncut american films in europe.:-/
I've had X die a hundred horrible deaths. If the machine isn't networked, it may as well have taken linux down, too, because there's no way to fix it except ctrl-alt-del. (When X is really gone ctrl-alt-backspace, etc. don't work.)
Though it is nice that the filesystems gets unmounted, it's still pretty bogus that it leaves the machine in an unusable state.
The last time this happen, I was running RedHat 6.0 & the/home partition filled up while editing a large image in the gimp. The gimp swap file in ~/.gimp choked on write, which led to an endless cascade of dialog boxes reporting the error. After a few thousand (X being unresponsive during this whole processes), X died completely.
One might hope that errors in apps wouldn't bring down the X server....
I wonder how AcceleratedX holds up in this condition. Anyone want to try this w/the gimp on AccelX? or code some other app to create windows until something chokes?
>Our founding fathers believed in God and this country was based on God. > All you have to do is take out that pocket full of change and read what it says
DOH! The founding fathers didn't put that on our coins. Revisionist christians did that in this century.
Instead of reading you change, read the works of the founding fathers. You'll get a decidedly different picture.
By every objective measure we have of the performance of public schools (granted, the tools are fairly crude standardized tests: ICTM, etc), they are doing a better job than they ever have. These test results are widely available.
This is particularly amusing when conservatives keep moaning about how we need more standardized tests. All the results contradict what they say about the public schools.
As someone who escaped from Kansas, I can report first hand that not "changing their curriculum" means basically ignoring evolution, which is also true of most midwestern & southern states. Everyone I know from the midwest that knows the theory learned it on their own time.
Evolution is given a cursory one or two days, and not exactly talked about in great detail.
One doesn't have to look far to see how the Board of Education got its odd ideas.
There is an article in Education Week (www.edweek.org, I believe) that talks about this trend a bit.
Well, no.;) There may be good arguments for a deity, but these are just false. It is not agreed that life can not be created from non-life. The most widely supported (by evidence) theories of life propose exactly that.
The other statements are similarly false (e.g. there is a progression from a common origin, both in the fossil record, and in DNA).
I empathize with the printer problems. I've tried at various times to use different photo quality printers under linux. It's never worked. Even the gimp's specialized drivers haven't worked.
Currently I build images with the gimp, then transfer them to windows to print them. The print server is actually a linux box. I just can't generate the printer output from linux. The whole unix print model seems pretty geared toward high output, low quality print serving.
> According to their data, Word and Excel dominated the Mac market, where they don't control the OS, before they dominated the Windows market.
When I read that I immediately wondered about the competition in the two markets. The Mac market has always been smaller (less competition), AFAIK, so it's easier to capture a large portion of it. You have to beat one or two products, not a dozen.
In which case making a direct comparison doesn't work, and the data doesn't support their argument.
That said, I recall Word being pretty good compared to other word processors, back in the day. I suspect control of the o/s had more to do with how Word remained on top, rather than how it got there -- it certainly seems like crap today (bloated and unreliable), and even life-long windows users seem unhappy with it.
> Humans have way more genes and DNA than bacteria
DNA, yes. genes, no. humans have huge amounts of non-coding DNA, but most human genes have bacterial analogs & the difference in number of genes isn't nearly as great as the difference in amount of DNA. i.e. we have a lot of noise in our DNA (including things like reminants of extinct retro-viruses which have spliced themselves into our genome. pretty wacky.)
the "noise" can be used to measure raw mutation rates (since a mutation in the non-coding region doesn't alter the animal much, though possibly affecting structural properties of its DNA), and it can be used to verify ancestral relationships.
Unfortunately this is a very political issue, so even though we might agree that scientific methods are the way to approach this, not everyone agrees to this.
Particularly contentious is the difference between inquiry based learning and rote memorization. All evidence points to inquiry based methods being the most effective, since they are about understanding the basis of knowledge, rather than memorizing results without knowing what they are based on.
But when kids start asking questions, you suddenly have to field questions like "What is the evidence for God?" "What is the evidence for the creation story?" "What is the basis of our belief?"
Systems of belief which rely on there being one immutable literal interpretation of a text are necessarily hostile to methods of teaching which encourage students to challenge, reinterpret, and understand the basis of what they are taught -- even when those methods are demonstrably more effective.
In short, to convince people to use the results of research, you must also address the threat you are making to their belief system.
Actually I'm trained as a scientist and engineer, and I'm not a teacher. But I am familiar with some of the research. Whole-language is not "doodoo". You flattly don't know what you're talking about. How about some evidence? You might start with Kapan for some reviews of the research supporting it. You have no idea what's been going on in classrooms. Modern theories (as derived from data, rather than assumption) have not been widely adopted. Putting it in the curriculum document doesn't mean it has been implemented. Social promotion is not generally something supported by educators, but supported by rich parents who won't stand to have johnny fail. And what is your beef with outcome based success? Being outcome based precisely means that the student is held to a well defined standard! It precisely means that educators have to be able to measure how effective their teaching is. It means you have to define terms like "mastery" when you use them in your curriculum document, so that they have definate meanings & can't be glossed over. For example in a writing curriculum you might define mastery as being able to compose a well structured paragraph based on criteria you define in the document. When you test the students they have to actually write such a paragraph. In contrast the standardized testing paradigm (which is in fact implemented this way in California) is to have a bubble test where the students *NEVER* actually write a paragraph. Instead they fill in bubble questions about a paragraph -- a task which doesn't even use the same part of the brain as composition.
What non-fudgable non-fuzzy objective measure are you held to?:)
A review process works. Criterion based tests might work, though there are problems with doing this.
>Educators, and modern educational theory, have little credibility in the face of obviously lower eductaional levels in kids today. Standardized tests are a way of holding educators' feet to the fire, and this is needed.
It's hard just to enumerate the number of wrong assumptions in this paragraph. But here goes.:)
1) Modern education theory for the most part hasn't been implemented. School boards and state mandates tend to be very conservative. Even in cases where they are implemented, teachers are not trained in them, so classroom practice hasn't actually changed at all. Much of what is called "modern educational theory" in fact isn't so modern, and is supported by evidence going back decades (e.g. there's about 60 years of evidence supporting whole-language methods, which CA just dumped from its curriculum based on back-to-basics hysteria & a complete absence of data supporting the move).
2) Educational levels of kids are not obviously lower, and even if they were it would not give an accurate account of educators. I don't know what evidence you are using. What many use are various norm referenced test scores (such as the international TIMMS test), which, however, suffer from numerous errors, such as using invalid sample sets. Students were not tested randomly in all countries. Some allowed self-selection, resulting in only the best students being tested. Even if there were evidence of falling standards, our student body has *not* been constant. E.g. California has had a huge influx of Spanish speaking students. Whenever this happens there is an extra burden on educators. Doing as good a job as they did 20 years ago is not enough to get the same results, because the student body is starting with a different set of skills.
3) Regarding the need to hold educator's feet to the fire, this is neither the problem nor the best solution even if it were the problem. Apathetic educators is not remotely the issue. The issues are moving spanish speaking students into all-english classrooms, gross under-budgeting ($20 per student per year classroom budget, where I live; this includes photocopy costs), and an almost complete lack of job training for teachers. For example teachers are seldom trained in implementing curriculum standards passed by the state.
Even if it were the case that apathetic educators were the issue, of all the ways of engaging them (positive reinforcement such as merit pay, negative reinforcement, or involving the teacher more in curriculum & assessment issues), negative reinforcement is dead last in effectiveness. By far a better approach would be to train and involve teachers in questions of curriculum and assessment. When they learn to document their own progress, it will mean much more than getting a single number at the end of the year that is somehow supposed to tell them how they're doing. Being involved in it gives a better understanding of what they can change, and it might actually make it interesting.
4) Bad standards are worse than no standards. Norm referenced tests nearly always lead to bad standards. If there are no standards, at least the good teachers are *allowed* to do a good job. If you implement a bad standard, you will only drive the good teachers away. Norm referenced tests are quite a big topic, which I don't have the time to go into here completely. Briefly, due to the renormalization process, test questions which can be answered by more than 60% of students are cut from the test. What is left are questions that are 1) not on the curriculum, 2) ambiguous or otherwise badly written, 3) obscure, or 4) beyond the skill level of the students is question.
In short they are not what we are interested in, and as a merit system they would establish a bad standard.
This might hold for private universities that get no tax dollars. But if you're getting tax money, you have an obligation to fairly represent the community.
Which means if you're trying to measure math ability to determine admissions, you better measure math ability, not math ability + culture.
For private institutions, it's less clear, though it would obviously be against their best interest to make the wrong measurement -- to reject those with better math skills based on a bad measurement.
There is a free market for tests. Unfortunately they are frequently chosen for the wrong reasons, since few people actually investigate what makes a meaningful test, and instead assume that every test measures what it claims to measure.
>Can't single kids out for special treatment and all that.
The problem with this is every parent wants it, and the special treatment is more causative of future success than indicative of previous success. If *any* student is pulled out for closer one-on-one work, they will do much better. Allocating those resources only for those already at the top of the class is hardly a good strategy.
But kids definately shouldn't be bored in class. A good teacher can arrange more advanced and independent work by kids who don't need hand-holding in some subject. We would do better to train teachers in keeping all the kids engaged, than to allocate special resources for those already on top.
What tends to happen is the rich parents insist on special treatment for their kids.
Well, actually new building are important. As long as it's not just new gyms, which is what tends to happen. Run down buildings can be very difficult places to learn, since it's difficult to focus on anything beyond how uncomfortable you are.
This is particularly a problem in hot climates, where you end up with a classroom of kids sweating on their desks & not terribly interested in anything except getting out of the building.
I'd also add more training of teachers. E.g. when there is new curriculum, you need to train teachers how to use it effectively -- same as tech people get job training. We have to invest a bit in our human resources.
When people say they don't accept "Darwinist evolution", they generally mean they go more for the theories of Gould than those of Dawkins, not that they have evidence against common origin via a mutable genetic material.
You'd have to overturn most of modern biology to refute the latter.
I agree with much of what you're saying, but this?
>there is no way that Newton's theory will ever be falsified.
All scientific knowledge is tentative. We reach the point where the chances of refuting a theory are essentially zero (as with evolution), but never precisely zero. There's no way to positively prove a theory is correct, even when there's so much evidence there is no reasonable doubt about it.
I'm currently reading "Skeptics and True Believers", which has some interesting commentary on this & other issues of science & belief (from a strongly skeptical perspective).
Again, they *have* to do this. Norm referenced tests are all about ranking students. If you fail to renormalize, the curve gets clipped on one end. With many more people taking the SAT, the curve shifts down. You start clipping the bottom. So you change the test to move the average back to 500.
All norm referenced tests do this. This is not political, it is the definition of a norm referenced test.
>nothing says that the academic performance among people of all ethnicities is equal, and that the SAT should confirm that
I wasn't arguing for that. Rather, that there is the result I already mentioned, and other results from norm referenced testing suggesting cultural biases unrelated to the material supposedly being tested. We hope the test measures mastery of a subject. However as far as we can tell it measures culture as well.
When all of this norm referenced testing started, it was generally held that cultural influences would overwhelm any comparisons outside a single classroom. Today we commonly assume that we can use it to meaningfully compare students in different countries, without, however, any evidence that this is true.
The assumption has changed, the data hasn't.
The claim that test results are vaild within a single classroom is much more well-grounded, because it eliminates a huge number of variables. Likewise, it is much more well-grounded to use the results within a single ethnic group than between them.
In the absence of evidence, prudence in interpreting tests results is called for.
>it is just a single factor and not one for prospective collegians to obsess about.
Slashdot Mirror
>The U.S. Surgeon General was recently asked to study the effects of media violence in children. He said they already did. The results are virtually unanimous: of 1000 studies, only 18 found no link between media violence and violence in youth.
I noted in a review of this data that they had failed to factor out things like time spent with parent. E.g. if kids do *anything* (say, watch "I love Lucy") to the exclusion of spending time with role models, they have a greater risk for violence and other anti-social behavior. It's very easy to get a correlation with "violence" which is actually merely a correlation with time in front of the TV & other baby-sitters instead of time with family & friends.
>I don't remember who stated that an X crash leaves Linux in an unusable state, but it's completely false
You're dead wrong. You assume that the X hot keys are still working, that the machine is networked, or that you have a 2.2 kernel with the sysrq keys compiled in & sysrq K actually manages to get X unstuck.
None of these are necessarily true. In particular, it's absurd to say "well, just buy two computers, so when X dies you can telnet in from the other one and fix it!" yeah, right. X shouldn't leave the machine stuck in a state where you *need another machine* to fix it.
I just ran another X hanging program (gnuplot a very large data file). When it locks, the ctrl-alt keys are *not* working. I do have sysrq, and sysrq K appears to have killed X (assuming it didn't die on its own) & since I have xdm running it was respawned.
However the ctrl-alt keys still don't work & in fact lock up X, now. ctrl-alt-f1 freezes the display. and following this sysrq K no longer works.
only sol'n is reboot or log in remotely (since i'm at work & have a dozen machines on the network). killing X via remote login, or going to run level 3 still doesn't get the consoles back. hot keys still don't work.
this is some pretty serious pathology for something as simple as "plot 'data'".
>we don't have laws like many European countries limiting things like how many foriegn movies our theaters can show.
:-/
No, in the US we have this cool rating system so tweaked that frequently we can't show uncut american made films in america, though you *can* show the uncut american films in europe.
Looks like maybe sysrq K will do it. I'll have to try this next time X totally dies.
This assumes that X is still alive enough to grab that hotkey, which isn't always the case.
If it's not, all you have are the kernel hotkeys, which don't get you much further than ctrl-alt-del, or the sysrq's.
I don't think any of those will get you to a console.
Anyone know for sure?
I've had X die a hundred horrible deaths. If the machine isn't networked, it may as well have taken linux down, too, because there's no way to fix it except ctrl-alt-del. (When X is really gone ctrl-alt-backspace, etc. don't work.)
/home partition filled up while editing a large image in the gimp. The gimp swap file in ~/.gimp choked on write, which led to an endless cascade of dialog boxes reporting the error. After a few thousand (X being unresponsive during this whole processes), X died completely.
Though it is nice that the filesystems gets unmounted, it's still pretty bogus that it leaves the machine in an unusable state.
The last time this happen, I was running RedHat 6.0 & the
One might hope that errors in apps wouldn't bring down the X server....
I wonder how AcceleratedX holds up in this condition. Anyone want to try this w/the gimp on AccelX? or code some other app to create windows until something chokes?
>Our founding fathers believed in God and this country was based on God.
> All you have to do is take out that pocket full of change and read what it says
DOH! The founding fathers didn't put that on our coins. Revisionist christians did that in this century.
Instead of reading you change, read the works of the founding fathers. You'll get a decidedly different picture.
By every objective measure we have of the performance of public schools (granted, the tools are fairly crude standardized tests: ICTM, etc), they are doing a better job than they ever have. These test results are widely available.
This is particularly amusing when conservatives keep moaning about how we need more standardized tests. All the results contradict what they say about the public schools.
As someone who escaped from Kansas, I can report first hand that not "changing their curriculum" means basically ignoring evolution, which is also true of most midwestern & southern states. Everyone I know from the midwest that knows the theory learned it on their own time.
Evolution is given a cursory one or two days, and not exactly talked about in great detail.
One doesn't have to look far to see how the Board of Education got its odd ideas.
There is an article in Education Week (www.edweek.org, I believe) that talks about this trend a bit.
Well, no. ;) There may be good arguments for a deity, but these are just false. It is not agreed that life can not be created from non-life. The most widely supported (by evidence) theories of life propose exactly that.
The other statements are similarly false (e.g. there is a progression from a common origin, both in the fossil record, and in DNA).
b.c.
what is SPA?
I empathize with the printer problems. I've tried at various times to use different photo quality printers under linux. It's never worked. Even the gimp's specialized drivers haven't worked.
Currently I build images with the gimp, then transfer them to windows to print them. The print server is actually a linux box. I just can't generate the printer output from linux. The whole unix print model seems pretty geared toward high output, low quality print serving.
> According to their data, Word and Excel dominated the Mac market, where they don't control the OS, before they dominated the Windows market.
When I read that I immediately wondered about the competition in the two markets. The Mac market has always been smaller (less competition), AFAIK, so it's easier to capture a large portion of it. You have to beat one or two products, not a dozen.
In which case making a direct comparison doesn't work, and the data doesn't support their argument.
That said, I recall Word being pretty good compared to other word processors, back in the day. I suspect control of the o/s had more to do with how Word remained on top, rather than how it got there -- it certainly seems like crap today (bloated and unreliable), and even life-long windows users seem unhappy with it.
> Humans have way more genes and DNA than bacteria
DNA, yes. genes, no. humans have huge amounts of non-coding DNA, but most human genes have bacterial analogs & the difference in number of genes isn't nearly as great as the difference in amount of DNA. i.e. we have a lot of noise in our DNA (including things like reminants of extinct retro-viruses which have spliced themselves into our genome. pretty wacky.)
the "noise" can be used to measure raw mutation rates (since a mutation in the non-coding region doesn't alter the animal much, though possibly affecting structural properties of its DNA), and it can be used to verify ancestral relationships.
Unfortunately this is a very political issue, so even though we might agree that scientific methods are the way to approach this, not everyone agrees to this.
Particularly contentious is the difference between inquiry based learning and rote memorization. All evidence points to inquiry based methods being the most effective, since they are about understanding the basis of knowledge, rather than memorizing results without knowing what they are based on.
But when kids start asking questions, you suddenly have to field questions like "What is the evidence for God?" "What is the evidence for the creation story?" "What is the basis of our belief?"
Systems of belief which rely on there being one immutable literal interpretation of a text are necessarily hostile to methods of teaching which encourage students to challenge, reinterpret, and understand the basis of what they are taught -- even when those methods are demonstrably more effective.
In short, to convince people to use the results of research, you must also address the threat you are making to their belief system.
Actually I'm trained as a scientist and engineer, and I'm not a teacher. But I am familiar with some of the research. Whole-language is not "doodoo". You flattly don't know what you're talking about. How about some evidence? You might start with Kapan for some reviews of the research supporting it. You have no idea what's been going on in classrooms. Modern theories (as derived from data, rather than assumption) have not been widely adopted. Putting it in the curriculum document doesn't mean it has been implemented. Social promotion is not generally something supported by educators, but supported by rich parents who won't stand to have johnny fail. And what is your beef with outcome based success? Being outcome based precisely means that the student is held to a well defined standard! It precisely means that educators have to be able to measure how effective their teaching is. It means you have to define terms like "mastery" when you use them in your curriculum document, so that they have definate meanings & can't be glossed over. For example in a writing curriculum you might define mastery as being able to compose a well structured paragraph based on criteria you define in the document. When you test the students they have to actually write such a paragraph. In contrast the standardized testing paradigm (which is in fact implemented this way in California) is to have a bubble test where the students *NEVER* actually write a paragraph. Instead they fill in bubble questions about a paragraph -- a task which doesn't even use the same part of the brain as composition.
What non-fudgable non-fuzzy objective measure are you held to? :)
:)
A review process works. Criterion based tests might work, though there are problems with doing this.
>Educators, and modern educational theory, have little credibility in the face of obviously lower eductaional levels in kids today. Standardized tests are a way of holding educators' feet to the fire, and this is needed.
It's hard just to enumerate the number of wrong assumptions in this paragraph. But here goes.
1) Modern education theory for the most part hasn't been implemented. School boards and state mandates tend to be very conservative. Even in cases where they are implemented, teachers are not trained in them, so classroom practice hasn't actually changed at all. Much of what is called "modern educational theory" in fact isn't so modern, and is supported by evidence going back decades (e.g. there's about 60 years of evidence supporting whole-language methods, which CA just dumped from its curriculum based on back-to-basics hysteria & a complete absence of data supporting the move).
2) Educational levels of kids are not obviously lower, and even if they were it would not give an accurate account of educators. I don't know what evidence you are using. What many use are various norm referenced test scores (such as the international TIMMS test), which, however, suffer from numerous errors, such as using invalid sample sets. Students were not tested randomly in all countries. Some allowed self-selection, resulting in only the best students being tested. Even if there were evidence of falling standards, our student body has *not* been constant. E.g. California has had a huge influx of Spanish speaking students. Whenever this happens there is an extra burden on educators. Doing as good a job as they did 20 years ago is not enough to get the same results, because the student body is starting with a different set of skills.
3) Regarding the need to hold educator's feet to the fire, this is neither the problem nor the best solution even if it were the problem. Apathetic educators is not remotely the issue. The issues are moving spanish speaking students into all-english classrooms, gross under-budgeting ($20 per student per year classroom budget, where I live; this includes photocopy costs), and an almost complete lack of job training for teachers. For example teachers are seldom trained in implementing curriculum standards passed by the state.
Even if it were the case that apathetic educators were the issue, of all the ways of engaging them (positive reinforcement such as merit pay, negative reinforcement, or involving the teacher more in curriculum & assessment issues), negative reinforcement is dead last in effectiveness. By far a better approach would be to train and involve teachers in questions of curriculum and assessment. When they learn to document their own progress, it will mean much more than getting a single number at the end of the year that is somehow supposed to tell them how they're doing. Being involved in it gives a better understanding of what they can change, and it might actually make it interesting.
4) Bad standards are worse than no standards. Norm referenced tests nearly always lead to bad standards. If there are no standards, at least the good teachers are *allowed* to do a good job. If you implement a bad standard, you will only drive the good teachers away. Norm referenced tests are quite a big topic, which I don't have the time to go into here completely. Briefly, due to the renormalization process, test questions which can be answered by more than 60% of students are cut from the test. What is left are questions that are 1) not on the curriculum, 2) ambiguous or otherwise badly written, 3) obscure, or 4) beyond the skill level of the students is question.
In short they are not what we are interested in, and as a merit system they would establish a bad standard.
This might hold for private universities that get no tax dollars. But if you're getting tax money, you have an obligation to fairly represent the community.
Which means if you're trying to measure math ability to determine admissions, you better measure math ability, not math ability + culture.
For private institutions, it's less clear, though it would obviously be against their best interest to make the wrong measurement -- to reject those with better math skills based on a bad measurement.
There is a free market for tests. Unfortunately they are frequently chosen for the wrong reasons, since few people actually investigate what makes a meaningful test, and instead assume that every test measures what it claims to measure.
>Can't single kids out for special treatment and all that.
The problem with this is every parent wants it, and the special treatment is more causative of future success than indicative of previous success. If *any* student is pulled out for closer one-on-one work, they will do much better. Allocating those resources only for those already at the top of the class is hardly a good strategy.
But kids definately shouldn't be bored in class. A good teacher can arrange more advanced and independent work by kids who don't need hand-holding in some subject. We would do better to train teachers in keeping all the kids engaged, than to allocate special resources for those already on top.
What tends to happen is the rich parents insist on special treatment for their kids.
Well, actually new building are important. As long as it's not just new gyms, which is what tends to happen. Run down buildings can be very difficult places to learn, since it's difficult to focus on anything beyond how uncomfortable you are.
This is particularly a problem in hot climates, where you end up with a classroom of kids sweating on their desks & not terribly interested in anything except getting out of the building.
I'd also add more training of teachers. E.g. when there is new curriculum, you need to train teachers how to use it effectively -- same as tech people get job training. We have to invest a bit in our human resources.
When people say they don't accept "Darwinist evolution", they generally mean they go more for the theories of Gould than those of Dawkins, not that they have evidence against common origin via a mutable genetic material.
You'd have to overturn most of modern biology to refute the latter.
I agree with much of what you're saying, but this?
>there is no way that Newton's theory will ever be falsified.
All scientific knowledge is tentative. We reach the point where the chances of refuting a theory are essentially zero (as with evolution), but never precisely zero. There's no way to positively prove a theory is correct, even when there's so much evidence there is no reasonable doubt about it.
I'm currently reading "Skeptics and True Believers", which has some interesting commentary on this & other issues of science & belief (from a strongly skeptical perspective).
Again, they *have* to do this. Norm referenced tests are all about ranking students. If you fail to renormalize, the curve gets clipped on one end. With many more people taking the SAT, the curve shifts down. You start clipping the bottom. So you change the test to move the average back to 500.
All norm referenced tests do this. This is not political, it is the definition of a norm referenced test.
>Dunno about racial bias (is there a box that says, "What color is your skin?"), but cultural bias
Yeah. Thanks for that correction.
b.c.
>nothing says that the academic performance among people of all ethnicities is equal, and that the SAT should confirm that
I wasn't arguing for that. Rather, that there is the result I already mentioned, and other results from norm referenced testing suggesting cultural biases unrelated to the material supposedly being tested. We hope the test measures mastery of a subject. However as far as we can tell it measures culture as well.
When all of this norm referenced testing started, it was generally held that cultural influences would overwhelm any comparisons outside a single classroom. Today we commonly assume that we can use it to meaningfully compare students in different countries, without, however, any evidence that this is true.
The assumption has changed, the data hasn't.
The claim that test results are vaild within a single classroom is much more well-grounded, because it eliminates a huge number of variables. Likewise, it is much more well-grounded to use the results within a single ethnic group than between them.
In the absence of evidence, prudence in interpreting tests results is called for.
>it is just a single factor and not one for prospective collegians to obsess about.
I agree completely.
I think you credit them with more intelligence than they have. I don't think any of them are thinking that far in advance.
I'm curious what evidence you have that educational standards have dropped since the 60's.