Not to sound selfish again, but you understand that funds for education are limited, right? Just because a new version of a history textbook is better in some minor way doesn't mean the expense is worth it.
Approximations are vital for children anyway. Things don't have to be absolutely right to get the general message across.
Furthermore, the "re-understanding" in a subject like US History is usually politically motivated anyway, so it's even less important. They generally aren't about factual errors but simply changes in what the politicians would like to emphasize. Again, definitely not worth millions of dollars that could otherwise go to better teachers, field trips, or hell even new sports equipment.
No, it isnt. If you think it is please post the relevant statute that makes it so.
Umm... results of arbitration are legally enforceable. See the Arbitration Act of 1996. (http://www.legislation.gov.uk/ukpga/1996/23/contents)
Maybe you don't know how these sharia courts are set up in the UK. They operate as arbitration tribunals that both parties have agreed to.
The problem, as I pointed out and you seem to have skipped over, is that sharia courts are discriminatory, and that a person's freedoms in religion, lifestyle, etc, may cause the sharia court to become discriminatory at a later date. For instance, if you, a Muslim, rent an apartment from me, another Muslim and agree to arbitration by sharia.. and then you become Christian.. and then next year we have a dispute? Guess what, the court will discriminate against you. Your testimony won't count as much as mine, for instance.
So to me that's quite obviously not a legitimate arbitration. Do you see the problem?
Compare that to a normal court, where no matter what you do in your personal life, you will have the same rights and privileges in court.
There are some cases where can sign your common law rights away, but in many cases you cannot. It is English (or European) law that decides which rights you can sign away, not some religous geezer.
Yeah, and the problem with that is a lot of victims of these sharia courts don't know their rights. They hear idiots saying stuff like "yeah you agreed to this, it's like a contract" and don't know that many things may not be contractually enforceable. Like if your mullah says a young girl in an arranged marriage can't divorce her abusive husband.. well she may not know that in the West women have more rights and she actually can get a divorce.
For such reasons, the mere specter of legitimacy is harmful to apply to sharia courts. It makes far more sense to say that Islamic law is not a valid basis for arbitration, no matter what you agree to, because its tenets are inherently discriminatory. Just rules of arbitration under Jim Crow law would be illegal and unenforceable.. in fact if you advertised something like housing with a requirement to sign arbitration under Jim Crow law, that itself would probably be illegal. Why should sharia be legal? It's religious freedom gone too far.
Even if our understanding changes, how important is that? Certainly we don't send letters to everybody who learned from the flawed textbooks, apologizing and offering a summary of the changes. So it probably doesn't matter. Is it worth the expense of writing and buying new textbooks?
If you rely on the teacher, you're making all the kids dependent on the teacher's level of quality. A lot of successful students practice by doing extra unassigned problems from the book, and they can check the answers in the back. You *might* get a good teacher who facilitates the same thing, but you probably won't.
This argument is absurd on its face. We benefit as a society when we all have access to education.
That argument is absurd because we don't publicly fund everything that benefits society. You would have to show that free education for children is more important than, say, free food, free housing, free transportation, etc. To me those are more fundamental, and yet they're largely self-funded. Why not have education be the same? If you're very poor, welfare could provide education, but otherwise leave it up to the parents.
If you read jackbird's post, you'll realise that he wasn't talking about motivation of the parents, but opportunity. A parent may be highly motivated for their child to succeed, but has to work 2-3 jobs just to keep them fed. That parent WON'T be able to get time off, because they're in such crappy jobs, their bosses will quite happily threaten to fire them if they don't show up for work. Because they can. (Let's completely ignore school selection based on ability to pay fees in a "timely" manner.)
Motivation is still a big factor in the student's academic success, even with fewer opportunities. Also we're talking about public school choice, that's where you get to choose which school to send your kid to, but it's still free. As for how long it takes to sign up for school, taking time off, etc to me that's a really silly issue to focus on. When you move to the school district or your kids age into the system, you have to fill out the same paperwork, whether you use the default school or pick another one. It's no harder to pick another one than to pick the default.
She and my step-father were very involved in my education, but didn't have enough education to help me beyond parent-teacher interviews and emotional support. The only other thing that was in my favour was that the state education system for the most part didn't play favourites.
If your education system didn't play favorites, then you must have been very lucky in the school you were forced to go to. If you happened to go to a horrible school, you would have been screwed. It would have been better for you if your parents could pick a better school and send you there for free.
But by all means, let's jam those unlucky bastards in with the really vicious students, simply because they can't afford private schools.
Separate issue.. vicious kids shouldn't be in normal schools. It's the opposite of what you said earlier.. we allow 5% of the population to ruin it for the 50% who are stuck in lower income, rural, or otherwise limited schools.
Well it's not just people settling their differences outside of court, it's a legally enforceable arbitration process. So if one says "Eh, you know what, I changed my mind I don't like this sharia garbage" too bad.
But I thought contracts or other agreements that have elements of discrimination is unenforceable? For instance if you said "To rent my property you have to agree to arbitration by Jim Crow law" -- and you apply that equally to all people -- that's still discriminatory against blacks and it wouldn't be enforceable.
Sharia has even more problems because its discriminatory status can change over time. For instance if you agree at time X to sharia arbitration, then at time X+1 you leave Islam and become Christian, what happens? Christians are discriminated against under sharia, so are you still bound by the sharia decision?
An agreement isn't supposed to be enforceable if it violates your rights and any agreement based on sharia does that, since sharia restricts your basic freedoms like freedom of religion which are in effect at all times in all situations.
My favorite teacher, by far, was my 10th grade history teacher. He told hilarious stories about his own life and hilarious stories about history. He was really sarcastic to kids who had bad behavior, often he got the whole class to laugh at them. The bad students hated him, the good students loved him.
Now.. unfortunately, I can't speak to his effectiveness as a teacher in terms of getting people to learn more than they would with a different teacher. I just don't have the perspective. I did well in every subject, whether the teacher was good or bad.
The teacher I liked least was my physics teacher who was also the football coach. I went to a good (but public) school so my AP classes were all full. There were plenty of people on various sports teams in the AP classes as well, and they loved him. He was popular with most students. He flirted with the girls and they thought he was super handsome. He was pretty smart and had decent knowledge of physics. He wasn't afraid to say he didn't know something if you asked a tough question.
But I just didn't like him. He didn't like me very much either. We just had clashing personalities. I think he used his willingness to admit lack of knowledge as a defense mechanism and a cover for his laziness, for instance.
But.. maybe he was a better teacher. Who knows.
My worst teacher, by far, was for a personal finance elective. First of all the teacher was very young, like a fresh grad, and didn't know the material. Not only that, she didn't know the supporting material -- basic math, computer literacy, etc. Nobody liked the class, but the teacher was very pretty, so all the guys (including me) liked her, at least as a person. She quickly figured that out and basically stopped teaching the class. She would tell people answers during tests, not care about homework, and so on. She was more interested in flirting than anything else.
I think this just underscores how important objective standards are. I don't think teachers should be rated on what amounts to a popularity contest. I have no doubt that if there were a standardized test on personal finance, most people in that class would have failed it and the teacher would be revealed to be awful. The physics teacher.. who knows. The lesson I draw from him is that different teachers work more effectively for different students. The history teacher was just perfect for someone like me. The physics teacher was just perfect for someone who was into sports but also wanted to excel academically with a bit of joking around and a desire to avoid the deeper stuff.
With that in mind, I think teachers should not receive a single performance metric. It should be broken down into a score per cluster of students. The clusters should not be defined in advance, they should be selected automatically by an algorithm that groups similar students together by demographic data, performance data, socio-economic data, etc.
There's absolutely nothing wrong with a teacher who can't teach AP English, but is really good at handling kids with discipline problems. They have a place in the system. Likewise, someone who helps people who are already self-motivated excel is valuable. It's not like self-motivated people can take care of themselves in all situations.. the successful disciplinarian teacher might try to crush that spirit and make them worse off for it.
The system needs to become more nuanced, and the most realistic way I see of that happening is standardized testing and automated analysis.
Of course, those are mostly rhetorical questions. The answer to all of them is because, "then people won't vote for me".
Oh... I doubt that. You'd have the votes. What you'd also get are lawsuits though.
I totally agree with holding parents responsible for student achievement. However, the state of America with respect to education is like the state of America with respect to Afghanistan and Iraq 10 years ago. There was this... euphoria that if we could just liberate them, they would all be happy and they would love us, etc. Nowadays that sentiment has changed and people are saying "If they don't like us, screw them."
The only thing that can save our education system is the same kind of growth and awareness. I believe that teacher performance metrics are a step down that road... when people see that teachers get great scores for years, then get assigned to a "problem class" and their scores plummet, we'll start to have this national recognition that the kids are the problem. Then we'll start asking why the kids are the problem, we'll try some more stupid things, but eventually we'll see that it's the parents who influence them the most and are the most responsible.
There was an article here just the other day explaining one possible system that seemed to have great results at a Chicago area charter school -- small monetary fines. It was just for disciplinary problems, but it would be very interesting to extend that to low test scores.
I'm curious what numbers were used in his performance that made him look bad?
Seems like if the company cares about money, and is tracking the money, then your husband did better than the previous manager.. so what's the problem?
I think the yearly improvement metric would still work at those schools. The Mexican immigrant children who did badly on a test this year probably did as bad or worse the year before when they knew even less English. It's not like they're in the top 10% with 9th grade English and then just crater to the lowest 10% in 10th grade.
I agree that measuring success isn't simple, but on the other hand you're cherry picking here. I think a year-on-year metric would work at a lot of places, and so why not use it at a lot of places? The places where it doesn't work can use something else. If you don't care about measuring the success of Mexican migrant children, then don't measure it... stick with the current wonderful system for them.
I'm hopeful that we'll develop a better idea of what we can expect from kids in the situation you described as a result of teacher evaluations. One day people will finally think "maybe it's not the teacher, maybe it's the student... look at how teacher evaluation scores are crushed across the board when they have more than X% of these bad students." Then we can be more efficient in allocating time and resources to achieve the best societal goals for education, rather than the striving for the lowest common denominator.
I don't think there are any easy metrics because it's a hard problem. Of the ones you mentioned, student evaluation is the worst because there's no way to safeguard that. Test score deltas seems like a good idea, and your criticism of it is incredibly easy to solve - the delta should be on the numerical score, not the grade letter. But of course since it's a test, it also shares the "teaching the test" problem of the first idea. I think that's also easy to solve -- don't tell the teachers details of what's on the test. Yes, scores will plummet. That's okay, curve them back up.
Seriously, you can't "teach the test" if all you know is "this test is about US history from the Civil War to WWII."
Is there something wrong with that type of clustering? It's certainly to the advantage of students who aren't that good but have motivated parents. The peer effect will help them, and the parental involvement will give them a high chance of improving and excelling.
If you hamper school choice and reduce that effect, it's not going to help the bad students as much. Their parents will still be unmotivated, and the peer effect will be reduced since the good students are more diluted too. The only "benefit" is that you get to hide the school's low aggregate performance by mixing in some higher scores.
That sucks, but it's the only realistic option the school system had. If all your ESL teachers are tenured, and then changing market conditions demand a change in teaching staff, the school is screwed. They can't fire someone and they can't hire past their budget. So they resort to shenanigans to solve the problem, which is far more traumatic and damaging -- good luck for this guy to get another job as a teacher after being fired for "touching".
If it were easier for schools to lay off unneeded teachers, this never would have happened, the school would have gotten better results earlier, and your friend could have just moved on to another job where his skills were more appreciated.
1) Teachers that "teach the test" - as a result we have mediocre educational performance getting rewarded.
Teaching the test should not be possible... it's only because teachers are given the test, or so much detail that they basically know the test.
The point of testing is that you demonstrate a small sample of knowledge drawn from a much larger pool. If you have no clue what the test is about except "US History" then getting an "A" on that test shows you probably know your stuff. If the teacher already knows the small sample of knowledge that the test will represent, it's worthless.
Teachers penalized for things not under their control - For example, in a large district like Manhattan, if teachers in the high-crime inner-city schools are evaluated in the same pool as the teachers serving students who live on Park Avenue, those teachers will be at a fundamental disadvantage simply because their job is harder.
I agree but that's not a good example. These systems usually are designed to measure changes in student performance from one year to the next. There's absolutely no reason why the teacher's score wouldn't be just as good if: 1. He teaches a class of smart, self-motivated students who all did well last year and all do well this year 2. He teaches a class of uninterested, academically unmotivated students who all did badly last year and all did badly this year
And in fact, if he can "reach" just a handful of those bad students, maybe his score would be better than in the first scenario.
Now, where this idea falls down is an example like.. comparing all the teachers who teach ESL classes, but half the schools have more funding and have separate ESL specialists that help the teacher. That teacher will obviously be seen as more effective than the teachers who don't have that additional support, and that's not really fair. In fact it falls down whenever you have a situation where more than 1 teacher is responsible for the same subject knowledge at the same time. I'm not sure how to fix that in the models, except excluding those cases. (Realistically, most students and most classes are average.)
I said "Standardized testing is by no means perfect but at least it's repeatable, measurable, and objective." That's standardized testing, not the teacher's performance metric. But I agree that ideally the teacher performance should also be repeatable for it to have much value -- especially if one-off decisions are made like firing someone.
Ravitch said "A teacher who is rated effective one year may well be ineffective the next year, depending on which students are assigned to his or her class."
Depending on the students.. well one thing I've read in other articles is that their scores change when they are given a new set of students who are really different from the previous year. For instance, someone who suddenly is teaching English to mostly non-native English speakers may have a very low performance grade. Isn't it possible that in these cases where there are significant changes in the students, that the teacher *actually isn't effective* in teaching that kind of student? I don't think teachers are one size fits all. (Perhaps this is another reason teachers are afraid of these statistical models.. they reveal uncomfortable truths. What will people think if it turns out the numbers show one teacher is much more effective at teaching white students than black students, for instance?)
It's true that it wouldn't be fair to fire a teacher for something like that. In this case, it would make more sense to have them teach an advanced or honors class where their strengths come out, and have someone else teach the basic class. If nobody can teach the basic class and show good performance, then you need to hire someone who can and maybe get rid of one of your redundant teachers. Isn't that how it SHOULD work? As opposed to now, where teachers are put into these tough classes, don't do a good job, and then get cycled out of them to easier classes. (I'm not naive, obviously that would continue even with the new statistical models... just because we have hard numbers showing there is a problem doesn't mean teachers and principles can't work together to beat the system.)
If Ravitch can't explain it to you, I can't.
That's pretty arrogant. I understood her opinion just fine, as I understood yours. I don't need it "explained" differently, I just disagree with it.
Wait, when was I talking about social issues? But lets go ahead and go through your attempted deflection, starting with affirmative action. Which party was Richard Nixon from, again?
You were talking about left vs. right, that includes social issues as well as economic. Do you not know that??
Nixon was a Republican. Are you being stupid on purpose? The Democrats controlled both the House and Senate during Nixon's presidency. You do know how our government works right?
Just how do small-government conservatives justify Big Government Action in preventing abortion, anyway?
Hah, good question. So now your evidence that the Democrats are right wing, and the Republicans are "insane" is... that Republicans have become less right wing and more left wing? That they favor big government over small? That's actually showing the exact opposite of what you wanted to show.
Uh, missing the obvious point that there's obviously no comparison between testing a single cell destined for a trash can and live animal testing, and thus no hypocrisy in being fore the former and against the latter?
Your "obvious point" is invalid. The hypothesis was the left is anti-science on some issues including animal testing. You're like "but Republicans don't want to test on single cells." My point was that single cell testing is not a substitute for animal testing. I'm helpfully trying to show you why talking about single cells is meaningless and does not invalidate the original hypothesis. Get it?
Maybe you could see a nice proctologist in North Korea for help on getting your head out, and then you can see what "left wing" really looks like.
Okay, wow good argument! Idiot.
Now, how about the Democrats supporting indefinite detention, assassinations of American citizens, austerity over job creation, endless war with ever-increasing DOD budgets, [blah blah]
Oh my God, you mean Democrats occasionally take "right wing" actions? Wow, I don't think they ever did that in the past, like that time they didn't round up a bunch of Japanese Americans after Pearl Harbor. Oh wait, they did do that in the past.
You live in a dream world where the left wing is ideologically pure and anything you don't like doesn't count as left wing, even though it's done by left wing people who at the time are spouting left wing nonsense. I bet you think Stalin wasn't communist, Mao wasn't communist, the "National Socialist" Nazi party wasn't socialist, etc.
The problem is half of your argument (Republicans have moved even farther right and are now insane) is invalidated by the same logic. There are a ton of left-wing actions that Republicans have taken (as you mentioned, bigger government, government intrusion in private life, etc)... so by your logic, they aren't right wing at all anymore! In fact the Republicans are left wing!
Nice try chief.
Too bad the facts don't match your storyline - starting to get to be a pattern here. Nuclear power is obscenely expensive: fact. [usatoday.com]
Umm.. pay attention, I said the technology is not expensive, but the construction is.. and if you actually read your own article it says "Nine environmental groups plan a challenge in federal court in Washington" -- wow, I wonder what effect that has on cost? Just like I said!
The same people hop back and forth between regulating nuclear power and working for the companies they just regulated: fact. [fpif.org]
I didn't dispute that. It's not "bad" like you believe though. OBVIOUSLY experts from the field need to be involved with regulation. You're an idiot if you don't see that. Guess what, shocking news for you... doctors are involved in their own regulation and certification processes! Because it turns out life long politicians don't know shit about medicine! Oh, breaking news, people who have worked in the
That was a much more balanced and informative article than the previous one about Ms. Isaacson. Even so, the conclusion of the article is rather biased.
For instance, it says:
It is possible to have higher test scores and worse education. The scores tell us nothing about how well students can think, how deeply they understand history or science or literature or philosophy, or how much they love to paint or dance or sing, or how well prepared they are to cast their votes carefully or to be wise jurors.
"Possible" is a weasel word. Sure it's possible, but that doesn't mean that the average case tells us nothing about the student's education. I could just as easily say "It is possible that higher test scores indicate better education."
And here's the alternative to testing as presented in the article:
Of course, teachers should be evaluated. They should be evaluated by experienced principals and peers. No incompetent teacher should be allowed to remain in the classroom. Those who can’t teach and can’t improve should be fired.
Okay, do you really not see the problem with teachers and principals policing each other, with absolutely no objectivity and no direct, measurable responsibility towards the students?
How many teachers are going to give an honest critical evaluation of their buddy who teaches the class next door, and say, "Yeah, X has no business being a teacher, he should be fired. Man he's a nice guy though!"
So what you end up with is a bunch of teachers and principals who are friends, and don't really care about student performance -- on standardized tests or anything else.
Standardized testing is by no means perfect but at least it's repeatable, measurable, and objective. That's soooo much better than the current state of things that any problems with the occasional great teacher being fired are more than compensated for by improving the overall teaching pool. I don't think it's healthy to focus on corner cases so much when dealing with something as fluid as education.
Oh, also, the inevitable outcome of widespread standardized testing, and performance evaluations based on the results, is that people will have a more realistic of what students can achieve. Sure, right now the PC thing is to say "You're teaching these dirt poor kids who live in gangs, and you should be able to accomplish the same thing as a teacher in a rich suburb." But we all know that's a load of crap. There are a ton of kids who just won't make it, and would actually be better off in a non-college oriented education program. Something really basic. And eventually the performance metrics will be turned around and used on the students as well as the teachers.
The article says no highly developed nation rates teachers like this.. well I don't know about that, but I know for a fact that several highly developed nations rate STUDENTS like this. That will happen here too, and we will all benefit.. we just have to get the ball rolling by embracing objectivity and measurement over the touchy-feely standards in use today.
Regarding the details of how the score is used, I suggest you re-read the article. It says pretty clearly that it's used as a factor for tenure decisions only, not firing:
This may seem disconnected from reality, but it has real ramifications. Because of her 7th percentile, Ms. Isaacson was told in February that it was virtually certain that she would not be getting tenure this year. “My principal said that given the opportunity, she would advocate for me,” Ms. Isaacson said. “But she said don’t get your hopes up, with a 7th percentile, there wasn’t much she could do.”
The firing / not rehiring is speculation:
That’s not the only problem Ms. Isaacson’s 7th percentile has caused. If the mayor and governor have their way, and layoffs are no longer based on seniority but instead are based on the city’s formulas that scientifically identify good teachers, Ms. Isaacson is pretty sure she’d be cooked.
See, pure speculation. Think about it, how could the school system afford to fire 7% of all its teachers every single year? There will ALWAYS be a lowest 7% of teachers so it never ends. The reporter and teacher are just being naive with their fears.
Almost everybody in the education field who follows the data agrees (like Diane Ravitch) that the factor that is most associated with student test scores is family income.
In this statistical model, the teacher's value-add is based on the change in performance of the students from year to year. Unless a significant number of students have massive changes in economic status from year to year, I'd need to see some reasoning on why your point isn't already taken into account just based on that. Also the article specifically says "The Lab School has selective admissions" so even if they change status this is be a group of very high achieving poor students, in which case the argument about income affecting performance doesn't necessarily apply to this sub-group because they've already been selected for performance.
But really you're speculating too much here given that we have almost no information about the statistical model. For instance the article says it takes into account 32 variables, which I'm speculating may include stuff like "belongs to the federal free-and-reduced school lunch program"... we just don't know. We also don't know that the value-add is calculated linearly.. a 2.04 to 2.10 improvement (which might be more typical among poor students) might count more than a 3.80 to 3.86.. both of these factors would help address your point.
Just a note, the reason we don't know these very important details is that the reporter just waved his hands in the air and said "NOBODY could understand this, so I won't bother reporting any facts about the statistical model. Who needs facts when I have this one outlier to prove my case."
In my reading of the NYT article, that's what the reporter meant when he said that the teacher scores aren't averages.
Even if the underlying numbers are adjusted, the final numbers are still averages. You and I see that, right? Once the proficiency score is calculated, an average for the teacher's students is computed. That's really straightforward. Saying that averages of adjusted numbers aren't averages means that you don't believe in weighted averages or nonlinear averages (eg geometric mean) or handy little things like standard deviation which depends on the average of a sum of squares.
Remember when you told me "If you don't know what a confidence interval is, there's no point in my talking to you"... imagine if I had made such ridiculous basic mistakes, and my defense was "Well that's not a confidence interval even though it says it's a confidence interval. I won't bother explaining why because I don't understand it myself, only a genius would get it." Wouldn't you have laughed at me? Yet that's almost
I read the article and I think the statistical model is much less problematic than you did... and the article is very problematic itself.
First of all the formula is used to identify teachers who are eligible for tenure, not to make a decision about firing them. That was a hypothetical, and assumes there would be no changes or other safeguards before a decision to fire is made.
Second, I agree with the formula that a teacher with 2.5 years experience has no business getting tenure. Do you? Perhaps the very wide confidence interval has something to do with the small dataset available for this teacher compared to other teachers in the system. That makes sense to me.
Third, no offense, but for all the questioning you're doing of my knowledge of statistics, I wonder if you have the same standards for the journalist who wrote this article? Look what he says: "Wrong. These are not averages. For example, the department defines Ms. Isaacson’s 3.57 prior proficiency as “the average prior year proficiency rating of the students who contribute to a teacher’s value added score.”"
But of course they ARE averages, it's right there in the definition. Why on earth did the journalist say "Wrong these are not averages" and then give a definition that starts "the average..."? Does he not like the fact that students who didn't contribute to a teacher's proficiency last year weren't counted? Should such students all have a default proficiency of 0 (to inflate the current year's progress) or 4 (to undermine it)? Don't you think that calls into question the journalist's ability to judge that the statistical model is incorrect, or call it so complex that "only Good Will Hunting could understand it?"
Honestly, I'm not saying I understand the entire statistical model based on the brief treatment given to it in the article, but are you seriously confident in dismissing it on the word of one English teacher who extrapolated her score to mean she would be fired, and a journalist who thinks he knows what an average is but is actually wrong?
I want to address this little tidbit from the article as well:
You would think the Department of Education would want to replicate Ms. Isaacson — who has degrees from the University of Pennsylvania and Columbia — and sprinkle Ms. Isaacsons all over town.
That's a common sentiment, but you can't replicate dedication and talent, and it's pretty rare to find. Most people are average -- in any field, not just teachers.
And in such a situation, it's the average that matters. So let's say the teacher really does get fired, because this statistical model is designed to find better average teachers, not superstar teachers. I'd rather lose Ms. Isaacson through an error and gain the benefit of having a better overall teacher pool because many truly bad teachers were eliminated. Wouldn't you? We can't rely on filling the school system with superstar teachers, it just isn't going to happen.
The journalist (and you) are criticizing this statistical model on the basis of one error (which we don't even know is an error.. sometimes the best intentioned, most enthusiastic, most well liked teacher isn't all that good at actually teaching). Does that seem very rational to you? I bet you anything that if the journalist interviewed some other 7th percentile teachers, he would find more than a few bad apples.
There's a huge movement in religion to accept that there are many paths to the same idea. You hear that a LOT if you spend much time talking to religious people. It's also a stock answer for whether someone would go to hell even if they were never exposed to Religion X -- why no, of course not, if they lived good lives in their own way they were on the right path all along, even if they weren't "officially" Religion X.
True. But that's because, to paraphrase Bill Maher, the Democratic Party has moved into the right wing, and the GOP has moved into the insane asylum.
Yeah all those social issues where the Democrats have moved right. Like how they now oppose gay marriage, oppose affirmative action, oppose abortion, and endorse religion in schools and public spaces... wait, no.
It must be all those right wing economic policies like opposing unions, flattening out tax burdens, opposing welfare spending, cutting wasteful education spending... err.
You must have a very left-wing definition of what right-wing means.
A single cell that's going to go in the trash anyway, as they're taken from fertility clinics? Vs a living, breathing animal made up of billions of cells that can think and feel pain?
Uh, yes? There are plenty of things that need whole organisms to test on, not single cells. You don't apply makeup to a single cell to figure out what impact it has on the single cell's eyes, right? Good luck testing a heart valve replacement on a single cell.
Ah, let the outright sophistry commence. Nuclear power is opposed for entirely rational, scientific reasons
No it's not. It's opposed by people who think nuclear power = proliferation, even though countries are developing nuclear weapons all on their own without our help. And also by people who think nuclear power is super dangerous and will cause giant frogs to rampage through our cities, or kill everybody on the East coast in one big puff of smoke. Or who think that nuclear waste is this big mysterious unsolvable problem and eventually the whole country will be contaminated because the waste lasts for billions of years.
There are many reasons to oppose the technology of nuclear power. None of them are rational in today's world.
Nuclear power plants are insanely expensive to build and run compared to other, greener sources of power.
They're expensive because of the insurance and regulatory delays required to address the fears of the previous groups. All these little activist groups that sue every few weeks and hold up construction for years, when billions of dollars are tied up in loans and already accruing interest... that's why nuclear power has been dead for 30 years.
It's debatable whether a rational response to an irrational action means the entire thing is rational (which you seem to be saying). I think since the foundation is irrational, the rational tidbit makes no difference.
And then there's the biggest flaws in nuclear power: human avarice and hubris.
That makes no sense. How are human avarice and hubris more applicable to nuclear power than to hydroelectric, coal, or solar thermal, for instance? How is that a flaw in nuclear power, not a flaw of humanity?
Obama is a good example of a Democrat willing to throw science under the bus. But no example of someone from the "left", given the fact that he's moved farther to the right than Reagan.
Nobody's on the left on every issue, but Obama and the Democratic party in general is from the "left" overall.
I don't "loathe" standardized tests. They are a useful tool, but they only measure what they measure.
Maybe you don't loathe them but that's the impression given in many of these articles by the anti-testing crowd.
To me the problem is that the standardized tests aren't hard enough, and teachers are told exactly what will be on each test. It should not be possible to "teach to the test" because that defeats the theory of testing -- a small sample of questions that you can statistically link to knowledge of the wider subject.
We also need better training of teachers, better pay for teachers to attract better new teachers, better administrative observation and assessment of teachers, a shift in focus from lecture to project and problem-based learning and more.
Interesting, all of those ideas hinge on the idea that current teachers aren't good enough. But before you said "If you aren't a teacher, you wouldn't know that sometimes no matter how hard you work and how well you teach, you get a bunch of kids that doesn't score well on tests." Why the change?
I think teachers are good enough, and always have been good enough. And they get paid enough.. actually too much right now considering their benefits. The problem is the kids and their parents. We need ways to engage them, like the fines this school is giving out. That's a great way to engage parents whose children need the most attention. It's so much more direct than any scheme I've heard of, except the valiant hero teacher who personally reaches out to the parents of every troubled kid in their classes. And that typically doesn't last because it requires an enormous amount of energy on the teacher's part.
And don't be so enamored of those "neutral third parties" who are making these tests. They are in it for money and money alone.
Of course they want more tests to be given, but I don't see any evidence that they don't want the tests to be high quality and useful. The people who have ruined standardized tests are the teachers and administrators and policy makers who want students to look better than they are.
Slashdot Mirror
Not to sound selfish again, but you understand that funds for education are limited, right? Just because a new version of a history textbook is better in some minor way doesn't mean the expense is worth it.
Approximations are vital for children anyway. Things don't have to be absolutely right to get the general message across.
Furthermore, the "re-understanding" in a subject like US History is usually politically motivated anyway, so it's even less important. They generally aren't about factual errors but simply changes in what the politicians would like to emphasize. Again, definitely not worth millions of dollars that could otherwise go to better teachers, field trips, or hell even new sports equipment.
No, it isnt. If you think it is please post the relevant statute that makes it so.
Umm... results of arbitration are legally enforceable. See the Arbitration Act of 1996. (http://www.legislation.gov.uk/ukpga/1996/23/contents)
Maybe you don't know how these sharia courts are set up in the UK. They operate as arbitration tribunals that both parties have agreed to.
The problem, as I pointed out and you seem to have skipped over, is that sharia courts are discriminatory, and that a person's freedoms in religion, lifestyle, etc, may cause the sharia court to become discriminatory at a later date. For instance, if you, a Muslim, rent an apartment from me, another Muslim and agree to arbitration by sharia.. and then you become Christian.. and then next year we have a dispute? Guess what, the court will discriminate against you. Your testimony won't count as much as mine, for instance.
So to me that's quite obviously not a legitimate arbitration. Do you see the problem?
Compare that to a normal court, where no matter what you do in your personal life, you will have the same rights and privileges in court.
There are some cases where can sign your common law rights away, but in many cases you cannot. It is English (or European) law that decides which rights you can sign away, not some religous geezer.
Yeah, and the problem with that is a lot of victims of these sharia courts don't know their rights. They hear idiots saying stuff like "yeah you agreed to this, it's like a contract" and don't know that many things may not be contractually enforceable. Like if your mullah says a young girl in an arranged marriage can't divorce her abusive husband.. well she may not know that in the West women have more rights and she actually can get a divorce.
For such reasons, the mere specter of legitimacy is harmful to apply to sharia courts. It makes far more sense to say that Islamic law is not a valid basis for arbitration, no matter what you agree to, because its tenets are inherently discriminatory. Just rules of arbitration under Jim Crow law would be illegal and unenforceable.. in fact if you advertised something like housing with a requirement to sign arbitration under Jim Crow law, that itself would probably be illegal. Why should sharia be legal? It's religious freedom gone too far.
Even if our understanding changes, how important is that? Certainly we don't send letters to everybody who learned from the flawed textbooks, apologizing and offering a summary of the changes. So it probably doesn't matter. Is it worth the expense of writing and buying new textbooks?
If you rely on the teacher, you're making all the kids dependent on the teacher's level of quality. A lot of successful students practice by doing extra unassigned problems from the book, and they can check the answers in the back. You *might* get a good teacher who facilitates the same thing, but you probably won't.
This argument is absurd on its face. We benefit as a society when we all have access to education.
That argument is absurd because we don't publicly fund everything that benefits society. You would have to show that free education for children is more important than, say, free food, free housing, free transportation, etc. To me those are more fundamental, and yet they're largely self-funded. Why not have education be the same? If you're very poor, welfare could provide education, but otherwise leave it up to the parents.
If you read jackbird's post, you'll realise that he wasn't talking about motivation of the parents, but opportunity. A parent may be highly motivated for their child to succeed, but has to work 2-3 jobs just to keep them fed. That parent WON'T be able to get time off, because they're in such crappy jobs, their bosses will quite happily threaten to fire them if they don't show up for work. Because they can. (Let's completely ignore school selection based on ability to pay fees in a "timely" manner.)
Motivation is still a big factor in the student's academic success, even with fewer opportunities. Also we're talking about public school choice, that's where you get to choose which school to send your kid to, but it's still free. As for how long it takes to sign up for school, taking time off, etc to me that's a really silly issue to focus on. When you move to the school district or your kids age into the system, you have to fill out the same paperwork, whether you use the default school or pick another one. It's no harder to pick another one than to pick the default.
She and my step-father were very involved in my education, but didn't have enough education to help me beyond parent-teacher interviews and emotional support. The only other thing that was in my favour was that the state education system for the most part didn't play favourites.
If your education system didn't play favorites, then you must have been very lucky in the school you were forced to go to. If you happened to go to a horrible school, you would have been screwed. It would have been better for you if your parents could pick a better school and send you there for free.
But by all means, let's jam those unlucky bastards in with the really vicious students, simply because they can't afford private schools.
Separate issue.. vicious kids shouldn't be in normal schools. It's the opposite of what you said earlier.. we allow 5% of the population to ruin it for the 50% who are stuck in lower income, rural, or otherwise limited schools.
Well it's not just people settling their differences outside of court, it's a legally enforceable arbitration process. So if one says "Eh, you know what, I changed my mind I don't like this sharia garbage" too bad.
But I thought contracts or other agreements that have elements of discrimination is unenforceable? For instance if you said "To rent my property you have to agree to arbitration by Jim Crow law" -- and you apply that equally to all people -- that's still discriminatory against blacks and it wouldn't be enforceable.
Sharia has even more problems because its discriminatory status can change over time. For instance if you agree at time X to sharia arbitration, then at time X+1 you leave Islam and become Christian, what happens? Christians are discriminated against under sharia, so are you still bound by the sharia decision?
An agreement isn't supposed to be enforceable if it violates your rights and any agreement based on sharia does that, since sharia restricts your basic freedoms like freedom of religion which are in effect at all times in all situations.
My favorite teacher, by far, was my 10th grade history teacher. He told hilarious stories about his own life and hilarious stories about history. He was really sarcastic to kids who had bad behavior, often he got the whole class to laugh at them. The bad students hated him, the good students loved him.
Now.. unfortunately, I can't speak to his effectiveness as a teacher in terms of getting people to learn more than they would with a different teacher. I just don't have the perspective. I did well in every subject, whether the teacher was good or bad.
The teacher I liked least was my physics teacher who was also the football coach. I went to a good (but public) school so my AP classes were all full. There were plenty of people on various sports teams in the AP classes as well, and they loved him. He was popular with most students. He flirted with the girls and they thought he was super handsome. He was pretty smart and had decent knowledge of physics. He wasn't afraid to say he didn't know something if you asked a tough question.
But I just didn't like him. He didn't like me very much either. We just had clashing personalities. I think he used his willingness to admit lack of knowledge as a defense mechanism and a cover for his laziness, for instance.
But.. maybe he was a better teacher. Who knows.
My worst teacher, by far, was for a personal finance elective. First of all the teacher was very young, like a fresh grad, and didn't know the material. Not only that, she didn't know the supporting material -- basic math, computer literacy, etc. Nobody liked the class, but the teacher was very pretty, so all the guys (including me) liked her, at least as a person. She quickly figured that out and basically stopped teaching the class. She would tell people answers during tests, not care about homework, and so on. She was more interested in flirting than anything else.
I think this just underscores how important objective standards are. I don't think teachers should be rated on what amounts to a popularity contest. I have no doubt that if there were a standardized test on personal finance, most people in that class would have failed it and the teacher would be revealed to be awful. The physics teacher.. who knows. The lesson I draw from him is that different teachers work more effectively for different students. The history teacher was just perfect for someone like me. The physics teacher was just perfect for someone who was into sports but also wanted to excel academically with a bit of joking around and a desire to avoid the deeper stuff.
With that in mind, I think teachers should not receive a single performance metric. It should be broken down into a score per cluster of students. The clusters should not be defined in advance, they should be selected automatically by an algorithm that groups similar students together by demographic data, performance data, socio-economic data, etc.
There's absolutely nothing wrong with a teacher who can't teach AP English, but is really good at handling kids with discipline problems. They have a place in the system. Likewise, someone who helps people who are already self-motivated excel is valuable. It's not like self-motivated people can take care of themselves in all situations.. the successful disciplinarian teacher might try to crush that spirit and make them worse off for it.
The system needs to become more nuanced, and the most realistic way I see of that happening is standardized testing and automated analysis.
Of course, those are mostly rhetorical questions. The answer to all of them is because, "then people won't vote for me".
Oh... I doubt that. You'd have the votes. What you'd also get are lawsuits though.
I totally agree with holding parents responsible for student achievement. However, the state of America with respect to education is like the state of America with respect to Afghanistan and Iraq 10 years ago. There was this... euphoria that if we could just liberate them, they would all be happy and they would love us, etc. Nowadays that sentiment has changed and people are saying "If they don't like us, screw them."
The only thing that can save our education system is the same kind of growth and awareness. I believe that teacher performance metrics are a step down that road... when people see that teachers get great scores for years, then get assigned to a "problem class" and their scores plummet, we'll start to have this national recognition that the kids are the problem. Then we'll start asking why the kids are the problem, we'll try some more stupid things, but eventually we'll see that it's the parents who influence them the most and are the most responsible.
There was an article here just the other day explaining one possible system that seemed to have great results at a Chicago area charter school -- small monetary fines. It was just for disciplinary problems, but it would be very interesting to extend that to low test scores.
I'm curious what numbers were used in his performance that made him look bad?
Seems like if the company cares about money, and is tracking the money, then your husband did better than the previous manager.. so what's the problem?
I think the yearly improvement metric would still work at those schools. The Mexican immigrant children who did badly on a test this year probably did as bad or worse the year before when they knew even less English. It's not like they're in the top 10% with 9th grade English and then just crater to the lowest 10% in 10th grade.
I agree that measuring success isn't simple, but on the other hand you're cherry picking here. I think a year-on-year metric would work at a lot of places, and so why not use it at a lot of places? The places where it doesn't work can use something else. If you don't care about measuring the success of Mexican migrant children, then don't measure it... stick with the current wonderful system for them.
I'm hopeful that we'll develop a better idea of what we can expect from kids in the situation you described as a result of teacher evaluations. One day people will finally think "maybe it's not the teacher, maybe it's the student... look at how teacher evaluation scores are crushed across the board when they have more than X% of these bad students." Then we can be more efficient in allocating time and resources to achieve the best societal goals for education, rather than the striving for the lowest common denominator.
I don't think there are any easy metrics because it's a hard problem. Of the ones you mentioned, student evaluation is the worst because there's no way to safeguard that. Test score deltas seems like a good idea, and your criticism of it is incredibly easy to solve - the delta should be on the numerical score, not the grade letter. But of course since it's a test, it also shares the "teaching the test" problem of the first idea. I think that's also easy to solve -- don't tell the teachers details of what's on the test. Yes, scores will plummet. That's okay, curve them back up.
Seriously, you can't "teach the test" if all you know is "this test is about US history from the Civil War to WWII."
Is there something wrong with that type of clustering? It's certainly to the advantage of students who aren't that good but have motivated parents. The peer effect will help them, and the parental involvement will give them a high chance of improving and excelling.
If you hamper school choice and reduce that effect, it's not going to help the bad students as much. Their parents will still be unmotivated, and the peer effect will be reduced since the good students are more diluted too. The only "benefit" is that you get to hide the school's low aggregate performance by mixing in some higher scores.
That sucks, but it's the only realistic option the school system had. If all your ESL teachers are tenured, and then changing market conditions demand a change in teaching staff, the school is screwed. They can't fire someone and they can't hire past their budget. So they resort to shenanigans to solve the problem, which is far more traumatic and damaging -- good luck for this guy to get another job as a teacher after being fired for "touching".
If it were easier for schools to lay off unneeded teachers, this never would have happened, the school would have gotten better results earlier, and your friend could have just moved on to another job where his skills were more appreciated.
1) Teachers that "teach the test" - as a result we have mediocre educational performance getting rewarded.
Teaching the test should not be possible... it's only because teachers are given the test, or so much detail that they basically know the test.
The point of testing is that you demonstrate a small sample of knowledge drawn from a much larger pool. If you have no clue what the test is about except "US History" then getting an "A" on that test shows you probably know your stuff. If the teacher already knows the small sample of knowledge that the test will represent, it's worthless.
Teachers penalized for things not under their control - For example, in a large district like Manhattan, if teachers in the high-crime inner-city schools are evaluated in the same pool as the teachers serving students who live on Park Avenue, those teachers will be at a fundamental disadvantage simply because their job is harder.
I agree but that's not a good example. These systems usually are designed to measure changes in student performance from one year to the next. There's absolutely no reason why the teacher's score wouldn't be just as good if:
1. He teaches a class of smart, self-motivated students who all did well last year and all do well this year
2. He teaches a class of uninterested, academically unmotivated students who all did badly last year and all did badly this year
And in fact, if he can "reach" just a handful of those bad students, maybe his score would be better than in the first scenario.
Now, where this idea falls down is an example like.. comparing all the teachers who teach ESL classes, but half the schools have more funding and have separate ESL specialists that help the teacher. That teacher will obviously be seen as more effective than the teachers who don't have that additional support, and that's not really fair. In fact it falls down whenever you have a situation where more than 1 teacher is responsible for the same subject knowledge at the same time. I'm not sure how to fix that in the models, except excluding those cases. (Realistically, most students and most classes are average.)
I said "Standardized testing is by no means perfect but at least it's repeatable, measurable, and objective." That's standardized testing, not the teacher's performance metric. But I agree that ideally the teacher performance should also be repeatable for it to have much value -- especially if one-off decisions are made like firing someone.
Ravitch said "A teacher who is rated effective one year may well be ineffective the next year, depending on which students are assigned to his or her class."
Depending on the students.. well one thing I've read in other articles is that their scores change when they are given a new set of students who are really different from the previous year. For instance, someone who suddenly is teaching English to mostly non-native English speakers may have a very low performance grade. Isn't it possible that in these cases where there are significant changes in the students, that the teacher *actually isn't effective* in teaching that kind of student? I don't think teachers are one size fits all. (Perhaps this is another reason teachers are afraid of these statistical models.. they reveal uncomfortable truths. What will people think if it turns out the numbers show one teacher is much more effective at teaching white students than black students, for instance?)
It's true that it wouldn't be fair to fire a teacher for something like that. In this case, it would make more sense to have them teach an advanced or honors class where their strengths come out, and have someone else teach the basic class. If nobody can teach the basic class and show good performance, then you need to hire someone who can and maybe get rid of one of your redundant teachers. Isn't that how it SHOULD work? As opposed to now, where teachers are put into these tough classes, don't do a good job, and then get cycled out of them to easier classes. (I'm not naive, obviously that would continue even with the new statistical models... just because we have hard numbers showing there is a problem doesn't mean teachers and principles can't work together to beat the system.)
If Ravitch can't explain it to you, I can't.
That's pretty arrogant. I understood her opinion just fine, as I understood yours. I don't need it "explained" differently, I just disagree with it.
Wait, when was I talking about social issues? But lets go ahead and go through your attempted deflection, starting with affirmative action. Which party was Richard Nixon from, again?
You were talking about left vs. right, that includes social issues as well as economic. Do you not know that??
Nixon was a Republican. Are you being stupid on purpose? The Democrats controlled both the House and Senate during Nixon's presidency. You do know how our government works right?
Just how do small-government conservatives justify Big Government Action in preventing abortion, anyway?
Hah, good question. So now your evidence that the Democrats are right wing, and the Republicans are "insane" is... that Republicans have become less right wing and more left wing? That they favor big government over small? That's actually showing the exact opposite of what you wanted to show.
Uh, missing the obvious point that there's obviously no comparison between testing a single cell destined for a trash can and live animal testing, and thus no hypocrisy in being fore the former and against the latter?
Your "obvious point" is invalid. The hypothesis was the left is anti-science on some issues including animal testing. You're like "but Republicans don't want to test on single cells." My point was that single cell testing is not a substitute for animal testing. I'm helpfully trying to show you why talking about single cells is meaningless and does not invalidate the original hypothesis. Get it?
Maybe you could see a nice proctologist in North Korea for help on getting your head out, and then you can see what "left wing" really looks like.
Okay, wow good argument! Idiot.
Now, how about the Democrats supporting indefinite detention, assassinations of American citizens, austerity over job creation, endless war with ever-increasing DOD budgets, [blah blah]
Oh my God, you mean Democrats occasionally take "right wing" actions? Wow, I don't think they ever did that in the past, like that time they didn't round up a bunch of Japanese Americans after Pearl Harbor. Oh wait, they did do that in the past.
You live in a dream world where the left wing is ideologically pure and anything you don't like doesn't count as left wing, even though it's done by left wing people who at the time are spouting left wing nonsense. I bet you think Stalin wasn't communist, Mao wasn't communist, the "National Socialist" Nazi party wasn't socialist, etc.
The problem is half of your argument (Republicans have moved even farther right and are now insane) is invalidated by the same logic. There are a ton of left-wing actions that Republicans have taken (as you mentioned, bigger government, government intrusion in private life, etc)... so by your logic, they aren't right wing at all anymore! In fact the Republicans are left wing!
Nice try chief.
Too bad the facts don't match your storyline - starting to get to be a pattern here. Nuclear power is obscenely expensive: fact. [usatoday.com]
Umm.. pay attention, I said the technology is not expensive, but the construction is.. and if you actually read your own article it says "Nine environmental groups plan a challenge in federal court in Washington" -- wow, I wonder what effect that has on cost? Just like I said!
The same people hop back and forth between regulating nuclear power and working for the companies they just regulated: fact. [fpif.org]
I didn't dispute that. It's not "bad" like you believe though. OBVIOUSLY experts from the field need to be involved with regulation. You're an idiot if you don't see that. Guess what, shocking news for you... doctors are involved in their own regulation and certification processes! Because it turns out life long politicians don't know shit about medicine! Oh, breaking news, people who have worked in the
That was a much more balanced and informative article than the previous one about Ms. Isaacson. Even so, the conclusion of the article is rather biased.
For instance, it says:
It is possible to have higher test scores and worse education. The scores tell us nothing about how well students can think, how deeply they understand history or science or literature or philosophy, or how much they love to paint or dance or sing, or how well prepared they are to cast their votes carefully or to be wise jurors.
"Possible" is a weasel word. Sure it's possible, but that doesn't mean that the average case tells us nothing about the student's education. I could just as easily say "It is possible that higher test scores indicate better education."
And here's the alternative to testing as presented in the article:
Of course, teachers should be evaluated. They should be evaluated by experienced principals and peers. No incompetent teacher should be allowed to remain in the classroom. Those who can’t teach and can’t improve should be fired.
Okay, do you really not see the problem with teachers and principals policing each other, with absolutely no objectivity and no direct, measurable responsibility towards the students?
How many teachers are going to give an honest critical evaluation of their buddy who teaches the class next door, and say, "Yeah, X has no business being a teacher, he should be fired. Man he's a nice guy though!"
So what you end up with is a bunch of teachers and principals who are friends, and don't really care about student performance -- on standardized tests or anything else.
Standardized testing is by no means perfect but at least it's repeatable, measurable, and objective. That's soooo much better than the current state of things that any problems with the occasional great teacher being fired are more than compensated for by improving the overall teaching pool. I don't think it's healthy to focus on corner cases so much when dealing with something as fluid as education.
Oh, also, the inevitable outcome of widespread standardized testing, and performance evaluations based on the results, is that people will have a more realistic of what students can achieve. Sure, right now the PC thing is to say "You're teaching these dirt poor kids who live in gangs, and you should be able to accomplish the same thing as a teacher in a rich suburb." But we all know that's a load of crap. There are a ton of kids who just won't make it, and would actually be better off in a non-college oriented education program. Something really basic. And eventually the performance metrics will be turned around and used on the students as well as the teachers.
The article says no highly developed nation rates teachers like this.. well I don't know about that, but I know for a fact that several highly developed nations rate STUDENTS like this. That will happen here too, and we will all benefit.. we just have to get the ball rolling by embracing objectivity and measurement over the touchy-feely standards in use today.
Regarding the details of how the score is used, I suggest you re-read the article. It says pretty clearly that it's used as a factor for tenure decisions only, not firing:
This may seem disconnected from reality, but it has real ramifications. Because of her 7th percentile, Ms. Isaacson was told in February that it was virtually certain that she would not be getting tenure this year. “My principal said that given the opportunity, she would advocate for me,” Ms. Isaacson said. “But she said don’t get your hopes up, with a 7th percentile, there wasn’t much she could do.”
The firing / not rehiring is speculation:
That’s not the only problem Ms. Isaacson’s 7th percentile has caused. If the mayor and governor have their way, and layoffs are no longer based on seniority but instead are based on the city’s formulas that scientifically identify good teachers, Ms. Isaacson is pretty sure she’d be cooked.
See, pure speculation. Think about it, how could the school system afford to fire 7% of all its teachers every single year? There will ALWAYS be a lowest 7% of teachers so it never ends. The reporter and teacher are just being naive with their fears.
Almost everybody in the education field who follows the data agrees (like Diane Ravitch) that the factor that is most associated with student test scores is family income.
In this statistical model, the teacher's value-add is based on the change in performance of the students from year to year. Unless a significant number of students have massive changes in economic status from year to year, I'd need to see some reasoning on why your point isn't already taken into account just based on that. Also the article specifically says "The Lab School has selective admissions" so even if they change status this is be a group of very high achieving poor students, in which case the argument about income affecting performance doesn't necessarily apply to this sub-group because they've already been selected for performance.
But really you're speculating too much here given that we have almost no information about the statistical model. For instance the article says it takes into account 32 variables, which I'm speculating may include stuff like "belongs to the federal free-and-reduced school lunch program"... we just don't know. We also don't know that the value-add is calculated linearly.. a 2.04 to 2.10 improvement (which might be more typical among poor students) might count more than a 3.80 to 3.86.. both of these factors would help address your point.
Just a note, the reason we don't know these very important details is that the reporter just waved his hands in the air and said "NOBODY could understand this, so I won't bother reporting any facts about the statistical model. Who needs facts when I have this one outlier to prove my case."
In my reading of the NYT article, that's what the reporter meant when he said that the teacher scores aren't averages.
Even if the underlying numbers are adjusted, the final numbers are still averages. You and I see that, right? Once the proficiency score is calculated, an average for the teacher's students is computed. That's really straightforward. Saying that averages of adjusted numbers aren't averages means that you don't believe in weighted averages or nonlinear averages (eg geometric mean) or handy little things like standard deviation which depends on the average of a sum of squares.
Remember when you told me "If you don't know what a confidence interval is, there's no point in my talking to you"... imagine if I had made such ridiculous basic mistakes, and my defense was "Well that's not a confidence interval even though it says it's a confidence interval. I won't bother explaining why because I don't understand it myself, only a genius would get it." Wouldn't you have laughed at me? Yet that's almost
I read the article and I think the statistical model is much less problematic than you did... and the article is very problematic itself.
First of all the formula is used to identify teachers who are eligible for tenure, not to make a decision about firing them. That was a hypothetical, and assumes there would be no changes or other safeguards before a decision to fire is made.
Second, I agree with the formula that a teacher with 2.5 years experience has no business getting tenure. Do you? Perhaps the very wide confidence interval has something to do with the small dataset available for this teacher compared to other teachers in the system. That makes sense to me.
Third, no offense, but for all the questioning you're doing of my knowledge of statistics, I wonder if you have the same standards for the journalist who wrote this article? Look what he says: "Wrong. These are not averages. For example, the department defines Ms. Isaacson’s 3.57 prior proficiency as “the average prior year proficiency rating of the students who contribute to a teacher’s value added score.”"
But of course they ARE averages, it's right there in the definition. Why on earth did the journalist say "Wrong these are not averages" and then give a definition that starts "the average..."? Does he not like the fact that students who didn't contribute to a teacher's proficiency last year weren't counted? Should such students all have a default proficiency of 0 (to inflate the current year's progress) or 4 (to undermine it)? Don't you think that calls into question the journalist's ability to judge that the statistical model is incorrect, or call it so complex that "only Good Will Hunting could understand it?"
Honestly, I'm not saying I understand the entire statistical model based on the brief treatment given to it in the article, but are you seriously confident in dismissing it on the word of one English teacher who extrapolated her score to mean she would be fired, and a journalist who thinks he knows what an average is but is actually wrong?
I want to address this little tidbit from the article as well:
You would think the Department of Education would want to replicate Ms. Isaacson — who has degrees from the University of Pennsylvania and Columbia — and sprinkle Ms. Isaacsons all over town.
That's a common sentiment, but you can't replicate dedication and talent, and it's pretty rare to find. Most people are average -- in any field, not just teachers.
And in such a situation, it's the average that matters. So let's say the teacher really does get fired, because this statistical model is designed to find better average teachers, not superstar teachers. I'd rather lose Ms. Isaacson through an error and gain the benefit of having a better overall teacher pool because many truly bad teachers were eliminated. Wouldn't you? We can't rely on filling the school system with superstar teachers, it just isn't going to happen.
The journalist (and you) are criticizing this statistical model on the basis of one error (which we don't even know is an error.. sometimes the best intentioned, most enthusiastic, most well liked teacher isn't all that good at actually teaching). Does that seem very rational to you? I bet you anything that if the journalist interviewed some other 7th percentile teachers, he would find more than a few bad apples.
There's a huge movement in religion to accept that there are many paths to the same idea. You hear that a LOT if you spend much time talking to religious people. It's also a stock answer for whether someone would go to hell even if they were never exposed to Religion X -- why no, of course not, if they lived good lives in their own way they were on the right path all along, even if they weren't "officially" Religion X.
True. But that's because, to paraphrase Bill Maher, the Democratic Party has moved into the right wing, and the GOP has moved into the insane asylum.
Yeah all those social issues where the Democrats have moved right. Like how they now oppose gay marriage, oppose affirmative action, oppose abortion, and endorse religion in schools and public spaces... wait, no.
It must be all those right wing economic policies like opposing unions, flattening out tax burdens, opposing welfare spending, cutting wasteful education spending... err.
You must have a very left-wing definition of what right-wing means.
A single cell that's going to go in the trash anyway, as they're taken from fertility clinics? Vs a living, breathing animal made up of billions of cells that can think and feel pain?
Uh, yes? There are plenty of things that need whole organisms to test on, not single cells. You don't apply makeup to a single cell to figure out what impact it has on the single cell's eyes, right? Good luck testing a heart valve replacement on a single cell.
Ah, let the outright sophistry commence. Nuclear power is opposed for entirely rational, scientific reasons
No it's not. It's opposed by people who think nuclear power = proliferation, even though countries are developing nuclear weapons all on their own without our help. And also by people who think nuclear power is super dangerous and will cause giant frogs to rampage through our cities, or kill everybody on the East coast in one big puff of smoke. Or who think that nuclear waste is this big mysterious unsolvable problem and eventually the whole country will be contaminated because the waste lasts for billions of years.
There are many reasons to oppose the technology of nuclear power. None of them are rational in today's world.
Nuclear power plants are insanely expensive to build and run compared to other, greener sources of power.
They're expensive because of the insurance and regulatory delays required to address the fears of the previous groups. All these little activist groups that sue every few weeks and hold up construction for years, when billions of dollars are tied up in loans and already accruing interest... that's why nuclear power has been dead for 30 years.
It's debatable whether a rational response to an irrational action means the entire thing is rational (which you seem to be saying). I think since the foundation is irrational, the rational tidbit makes no difference.
And then there's the biggest flaws in nuclear power: human avarice and hubris.
That makes no sense. How are human avarice and hubris more applicable to nuclear power than to hydroelectric, coal, or solar thermal, for instance? How is that a flaw in nuclear power, not a flaw of humanity?
Obama is a good example of a Democrat willing to throw science under the bus. But no example of someone from the "left", given the fact that he's moved farther to the right than Reagan.
Nobody's on the left on every issue, but Obama and the Democratic party in general is from the "left" overall.
I don't "loathe" standardized tests. They are a useful tool, but they only measure what they measure.
Maybe you don't loathe them but that's the impression given in many of these articles by the anti-testing crowd.
To me the problem is that the standardized tests aren't hard enough, and teachers are told exactly what will be on each test. It should not be possible to "teach to the test" because that defeats the theory of testing -- a small sample of questions that you can statistically link to knowledge of the wider subject.
We also need better training of teachers, better pay for teachers to attract better new teachers, better administrative observation and assessment of teachers, a shift in focus from lecture to project and problem-based learning and more.
Interesting, all of those ideas hinge on the idea that current teachers aren't good enough. But before you said "If you aren't a teacher, you wouldn't know that sometimes no matter how hard you work and how well you teach, you get a bunch of kids that doesn't score well on tests." Why the change?
I think teachers are good enough, and always have been good enough. And they get paid enough.. actually too much right now considering their benefits. The problem is the kids and their parents. We need ways to engage them, like the fines this school is giving out. That's a great way to engage parents whose children need the most attention. It's so much more direct than any scheme I've heard of, except the valiant hero teacher who personally reaches out to the parents of every troubled kid in their classes. And that typically doesn't last because it requires an enormous amount of energy on the teacher's part.
And don't be so enamored of those "neutral third parties" who are making these tests. They are in it for money and money alone.
Of course they want more tests to be given, but I don't see any evidence that they don't want the tests to be high quality and useful. The people who have ruined standardized tests are the teachers and administrators and policy makers who want students to look better than they are.