The less-funding-for-lower-scores anti-pattern implemented by NCLB is especially ridiculous.
But like I said before, I think the solution is to accept that only qualitative metrics will be effective predictors of performance. The "quantitative" tests are really just as subjective in what they value.
I've always positively hated the tests in which a question will be so dumbed-down that none of the choices is strictly true, for example.
In California's new primary system, we all vote in the same open primary and the top two candidates appear on the general election ballot regardless of party. There is only one primary, which basically operates like a run-off.
In the Bay Area, in the first year of the system, this meant Democrat vs. Democrat, Democrat vs. Green or Democrat vs. Peace & Freedom.
A simple application of Bayes' Law illustrates that if a test is 99% sensitive and 99% specific, the actual probability of an accurate match is only 33%.
Not to mention the fact that the democracies which actually exist do not implement simple majority-rules tyrannies. They implement representative or hybrid representative / direct systems with significant checks on the power of majorities and minorities.
Convince your friends and family to support open primary, citizens' redistricting and national popular vote laws in your state (mine has all three), and most important, never vote tactically. If you want a smaller government, vote Libertarian. If you want a full employment economy, vote Green.
There is a choice, but you have to fight to make it a meaningful one.
The better choice was voting for a politician or party whose policies you support (even if they lose).
Thanks to California's open primaries and Citizens' Redistricting Committee I am 100% able to do this at the local and state level from 2012 on. I did vote tactically on national offices, except the presidency, where again I voted for a party with sound policies even though they were likely to lose.
"Metadata" sounds non-threatening to a public ignorant of the fact that the metadata consists in all calling records, including inbound / outbound numbers, time, duration and GPS locations.
I don't want the USA to be "like Europe" (a meaningless phrase if ever there was one). Medicare for all (as an example) wouldn't make the USA "like Europe," it would just make it healthier, wealthier and still uniquely American.
Under our system tax revenues absolutely do not operationally fund government spending. Actually, we have to count on them realizing they could be the next Beardo.
Except this time. Scalia may be a troll, but I would argue that Alito and Thomas are worse. I honestly prefer hateful scumbags to hateful authoritarian scumbags.
The Mathematics questions test of basic arithmetic. It appears designed to be relatively easy for good students. Some of the terminology might be unfamiliar to you if you went to US primary school before the mid-late 90's (or just can't remember fourth grade very well). There aren't a lot of ways to write multiple choice questions for fourth grade arithmetic...some of the word problems could be more interesting or "real world" based, I guess.
The Science questions likewise test elementary school science. This largely entails basic knowledge about the world, such as why the moon shines, why getting too much sun can be a bad thing or the germ theory of disease. As such it inherently consists of learning terminology which supports a basic conceptual framework for science at a level where it does overlap significantly with (adult) "common sense." Any test of this type of framework will be a test of the essential terminology of the framework. This continues to be true in science long after the elementary school level. University physics, chemistry and biology all have large bodies of linked concepts which are taught and expressed through a large, specialized terminology. Personally, I would like to see more experimentally focused questions, of which there are essentially none, but I acknowledge that this is because inadequate funding has destroyed the ability of most American schools to conduct experiments in class. These tests are also intended for international distribution, which further constrains what is reasonable. It's an non-ideal test for a non-ideal system, but it's not at all egregious, especially considering that these "easy for a good student" type exams are really 1 or 2 grades lower than labelled in real difficulty
Tests like these exist because there is a demand for "accountability" through quantitative and behavioral performance metrics. Their scale and desired uniformity require simple multiple choice exams, which in turn must test many small, concrete problems in using some common vocabulary. If you really want to know what somebody "understands" internally, there is no test or technology that can achieve that. Detailed evaluations of individual students written by their teachers are the only thing that could even come close. If it were up to me, that's what I would choose (and adequately fund). The college I went to had such Narrative Evaluations instead of grades (later relaxed into a hybrid system with grades for large quantitative classes); in my opinion it's a far superior system. The students even evaluate professors and TAs in turn, which are reviewed by the school and the teachers, and have a real impact.
Standardized, multiple choice exams have problems. They usually don't accurately reflect ability, discriminate against ELLs as well as students with anxiety or dyslexia and are generally vulnerable to gaming / test strategies. But most I've seen aren't what I'd call tests of memorization, except maybe the History component of the STAR exams.
Slashdot Mirror
The less-funding-for-lower-scores anti-pattern implemented by NCLB is especially ridiculous.
But like I said before, I think the solution is to accept that only qualitative metrics will be effective predictors of performance. The "quantitative" tests are really just as subjective in what they value.
I've always positively hated the tests in which a question will be so dumbed-down that none of the choices is strictly true, for example.
In California's new primary system, we all vote in the same open primary and the top two candidates appear on the general election ballot regardless of party. There is only one primary, which basically operates like a run-off.
In the Bay Area, in the first year of the system, this meant Democrat vs. Democrat, Democrat vs. Green or Democrat vs. Peace & Freedom.
Mod parent up, it's implied by Snowden's comments it was witnessing situations much like these hypotheticals that drove him to blow a whistle.
I like your "one problem at a time" theory, it's a really great way never to address any particular issue at hand.
And like many people, maybe you forget that "metadata" includes your GPS location.
A simple application of Bayes' Law illustrates that if a test is 99% sensitive and 99% specific, the actual probability of an accurate match is only 33%.
Especially 16 KHz, 64-bit VBR MP3s, or even lower quality, which is perfectly sufficient for recorded voice.
And in this case he also happens to be a professor of constitutional law. You can't make this shit up.
Not to mention the fact that the democracies which actually exist do not implement simple majority-rules tyrannies. They implement representative or hybrid representative / direct systems with significant checks on the power of majorities and minorities.
Convince your friends and family to support open primary, citizens' redistricting and national popular vote laws in your state (mine has all three), and most important, never vote tactically. If you want a smaller government, vote Libertarian. If you want a full employment economy, vote Green.
There is a choice, but you have to fight to make it a meaningful one.
The better choice was voting for a politician or party whose policies you support (even if they lose).
Thanks to California's open primaries and Citizens' Redistricting Committee I am 100% able to do this at the local and state level from 2012 on. I did vote tactically on national offices, except the presidency, where again I voted for a party with sound policies even though they were likely to lose.
Don't forget GPS location. If you can't turn off location sending for 911, you can't for the NSA either.
"Metadata" sounds non-threatening to a public ignorant of the fact that the metadata consists in all calling records, including inbound / outbound numbers, time, duration and GPS locations.
I don't want the USA to be "like Europe" (a meaningless phrase if ever there was one). Medicare for all (as an example) wouldn't make the USA "like Europe," it would just make it healthier, wealthier and still uniquely American.
"Empirically sound policy" != "like Europe."
Under our system tax revenues absolutely do not operationally fund government spending. Actually, we have to count on them realizing they could be the next Beardo.
I'm talking about economic and personal freedom
What your talking about is freedom for the (financially) mighty to oppress the small. You might as well dispense with your coded speech.
there is no place else in the world that has the economic opportunity and freedom that the US has
Maybe that was true once, but it has long been a commonplace that upwards social mobility is greater in Europe than in the United States.
Good point, there goes their "fragmentation" legal argument.
Usually that's caused by the max heap setting being too small (which was exactly the parent's point).
Except this time. Scalia may be a troll, but I would argue that Alito and Thomas are worse. I honestly prefer hateful scumbags to hateful authoritarian scumbags.
The Mathematics questions test of basic arithmetic. It appears designed to be relatively easy for good students. Some of the terminology might be unfamiliar to you if you went to US primary school before the mid-late 90's (or just can't remember fourth grade very well). There aren't a lot of ways to write multiple choice questions for fourth grade arithmetic...some of the word problems could be more interesting or "real world" based, I guess.
The Science questions likewise test elementary school science. This largely entails basic knowledge about the world, such as why the moon shines, why getting too much sun can be a bad thing or the germ theory of disease. As such it inherently consists of learning terminology which supports a basic conceptual framework for science at a level where it does overlap significantly with (adult) "common sense." Any test of this type of framework will be a test of the essential terminology of the framework. This continues to be true in science long after the elementary school level. University physics, chemistry and biology all have large bodies of linked concepts which are taught and expressed through a large, specialized terminology. Personally, I would like to see more experimentally focused questions, of which there are essentially none, but I acknowledge that this is because inadequate funding has destroyed the ability of most American schools to conduct experiments in class. These tests are also intended for international distribution, which further constrains what is reasonable. It's an non-ideal test for a non-ideal system, but it's not at all egregious, especially considering that these "easy for a good student" type exams are really 1 or 2 grades lower than labelled in real difficulty
Tests like these exist because there is a demand for "accountability" through quantitative and behavioral performance metrics. Their scale and desired uniformity require simple multiple choice exams, which in turn must test many small, concrete problems in using some common vocabulary. If you really want to know what somebody "understands" internally, there is no test or technology that can achieve that. Detailed evaluations of individual students written by their teachers are the only thing that could even come close. If it were up to me, that's what I would choose (and adequately fund). The college I went to had such Narrative Evaluations instead of grades (later relaxed into a hybrid system with grades for large quantitative classes); in my opinion it's a far superior system. The students even evaluate professors and TAs in turn, which are reviewed by the school and the teachers, and have a real impact.
Some things never change. Although in this case it's a student visa (assuming he's not a resident already)...
This (PDF) doesn't seem like a memorization based exam to me...
Nor does this one...
Standardized, multiple choice exams have problems. They usually don't accurately reflect ability, discriminate against ELLs as well as students with anxiety or dyslexia and are generally vulnerable to gaming / test strategies. But most I've seen aren't what I'd call tests of memorization, except maybe the History component of the STAR exams.
I guess you missed the part about him going to Cornell and working at Google?
Answer: ls --help | grep '\-l\s'
there is no absolutely "fair" way of dividing a large group into two mutually exclusive categories
Most insightful thing I've read all week.