For an absurd idea, it's generated a lot of research:

http://answers.google.com/answers/threadview/id...
http://epm.sagepub.com/cgi/content/abstract/42/...
http://findarticles.com/p/articles/mi_m0FCR/is_...
http://www1.istation.com/en/corpsite/research/p...
http://www.readnaturally.com/pdf/RFBATechnicalD...
http://etd.lib.ttu.edu/theses/available/etd-073...

There is no reason to believe that just because a standardised test measures something that the measure is valid. But, a standardised test can be validated, and if it is properly valid then of course it measures things other than how well students take standardised tests.

How many of you take a standardized test every day at work and are paid based on the results of that test?

Actually my husband does this about once a week, he's not paid on the results of his test, but the company's finanical rewards are tied to his results. He's an engineer.

Of the best and the brightest students we've had in the past ten to twenty years, many of them went into investment banking because they were taught for years that the most important measure of success is money.

I don't believe you. Hollywood movies and TV have been teaching for years that family and friends are far more important than money. I think the students went into investment banking because they decided, for themselves, that they wanted money, despite the propaganda of the media and schools.

Why? Is it because they had bad teachers? Is it because they were perhaps taught to work within a system and to not question it, particularly not to question it if it led to greater profits, the holy grail of our society?

Or, alternatively, it's because humans have a tendency to do something that produces rewards in the short-term, but every now and then turns out to be really risky. After all, humans often drive cars despite the risks involved in that.

As for teaching time – the results of Direct Instruction curriculum I discussed above shows that it's possible for a school to track the progress of every single kid at the school and adjust their practice based on it – this however is not just about teachers, it's about the whole of the school.

“I have followed your work for years in “The New Yorker” and have found your writing to be consistently interesting and intellectually thought-provoking, even when I disagree with your conclusions. A case in point is your latest article about education. I have recently spent the past week boning up on the literature related to what, if anything, can be used as a proxy to determine whether or not a potential teacher hire will be effective in the classroom…

So when I came across the following sentences in your article I did a double take:

“A group of researchers–Thomas J. Kane, an economist at Harvard's school of education; Douglas Staiger, an economist at Dartmouth; and Robert Gordon, a policy analyst at the Center for American Progress–have investigated whether it helps to have a teacher who has earned a teaching certification or a master's degree. Both are expensive, time-consuming credentials that almost every district expects teachers to acquire; neither makes a difference in the classroom. Test scores, graduate degrees, and certifications–as much as they appear related to teaching prowess–turn out to be about as useful in predicting success as having a quarterback throw footballs into a bunch of garbage cans.”

First of all, I'm sure you had to simply this section for the demands of an magazine article, but still, I don't think you do justice to the state of the art research on this subject. First of all, there is a bunch of research (including a recent NBER Working Paper by among others Kane and Staiger) which suggests there are some tools that can help screen for effective teachers. To quote from the abstract (“Can You Recognize An Effective Teacher When You Recruit One?”):

“…we administered an in-depth survey to new math teachers in New York City and collected this information on a number of non-traditional predictors of effectiveness including teaching specific content knowledge, cognitive ability, personality traits, feelings of self-efficacy, and scores on a commercially available teacher selection instrument. Individually, we find that only a few of these predictors have statistically significant relationships with student and teacher outcomes. However, when all of these variables are combined into two primary factors summarizing cognitive and non-cognitive teacher skills, we find that both factors have a modest and statistically significant relationship with student and teacher outcomes, particularly with student test scores.”

Other older studies show that in certain cases content knowledge can increase student performance (mostly for math and science and for high-school students as opposed to elementary school students) and even more intriguingly, the general literacy level of teachers can be a powerful predictor of their effectiveness with students (check out page 8 of the this booklet published by the National Council on Teacher Quality:

http://www.nctq.org/p/publications/docs/nctq_io...)

So I think you mislead your readers to suggest that there are no good predictors of teaching prowess. On the other hand, the conclusions you reach still make sense for most large school districts, given that “value-added” is still the gold standard for determining who will be a good teacher over their career. Therefore, the idea that we need to make it easier for lots of different types of students and/or mid-career professionals thinking of going into teaching to become teachers and then be willing to take the time and effort to weed out the bad apples and reward and nurture those individuals who show promise, is right on.

If teaching is in fact a talent, and we pay people according to talents and their rarity given a particular demand… suddenly good teachers are worth a lot of money. Not NFL quarterback money but…

The idea that a great teacher is worth 500,000 bucks a year is not something we are comfortable thinking or admitting, but if you want to see a lot of talented and smart people be motivated to do great things as teachers…

Much of the discussion here focuses on reforming schools and how to fix the teacher credentialing process, etc. These are good questions, but there are some far larger questions that Gladwell misses and that aren't discussed here as well.

Our students are in school longer than they've ever been before, why aren't their scores improving? Perhaps they should be in school less? No one ever asks that question because we need kids to be in school so their parents don't have to be responsible for them. Lets not forget that teachers, at a certain point, are nothing more than glorified babysitters. I am one, and the fact that I could lose my job pretty quickly if I didn't take attendance as opposed to the fact that I could show my students movies every day for years and very likely not get fired suggests as much.

Of the best and the brightest students we've had in the past ten to twenty years, many of them went into investment banking because they were taught for years that the most important measure of success is money. I-banking was a great place to make a lot of money. All these kids that scored off the charts on standardized tests and went to the best universities in the country just drove a giant financial machine off a cliff. Why? Is it because they had bad teachers? Is it because they were perhaps taught to work within a system and to not question it, particularly not to question it if it led to greater profits, the holy grail of our society?

Does anyone ask what we did wrong with these kids? They have high IQ's, they got straight A's, they got all the right stamps on their passport to financial success, but they completely missed a hundred huge clues that something was very wrong? Why? Were they not spending enough time in the office? Should they have had more math class and less gym class so they could really be prepared?

I agree with much of what's been stated here, the credentialing system is terrible and not worth the time and money spent on it, majoring in Education should be outlawed because most of it has almost nothing to do with teaching in the real world and being competent in the subject you plan to teach is far more relevant than a degree in Education, teachers ought to be compensated better so that you can encourage more talented and motivated people to enter the field and STAY in the field, all these things are true.

But there are larger problems that have to be addressed first. Why are high school kids in school at 7:30? It doesn't make any sense physiologically or psychologically, they'd be better off coming in at 9 and leaving at 2:30.

Is it rational to expect teachers to be able to adequately prepare for 4-5 classes a day every day with only 1-2 hours of prep time? Would any college professor agree to this? Would any manager agree to run 4-5 meetings a day with anywhere from 15-40 people who may or may not want to be there, and then be responsible to tracking the progress of each of those employees and adjusting practice based on that? And do that every day, every week for 180 days of the year? Of course not, they don't get paid enough to do that. So why on earth would they choose to do it as a teacher?

As long as we run our schools like factories where children progress down an assembly line according to bells that ring and we measure them by standardized tests that measure one form of intelligence, we will continue to destroy creativity and initiative in more than ninety percent of our students. As long as we pay teachers a pittance compared to professions with similar demands, we will continue to get a lackluster crowd of folks doing it with a few exceptions. As long as we think about schools as a way to get a certain product rather than a place to grow students into whatever they want/need to be, we will continue to fill the workplace and the world with a few bright successes and dump the rest into reject lots just like Detroit's has done with all the cars that don't pass inspection at the end of the line. As long as it is less expensive to run education that way, we will continue to do it.

The problem is that very soon we are going to have to pay the piper and very few people understand the scale of the problem or the enormous expense it will take to fix it. This financial meltdown is just the tip of the iceberg in comparison.

You also appear to assume that the relevant factor in education spending is the quality of the teacher, not the quality of the school. However, there is a lot of evidence that the quality of the school in terms of the principal and the quality of the students attending it matter greatly. See for example http://www.nwrel.org/scpd/esp/esp95toc.html Hedonic studies of housing prices imply that people are willing to spend more in at least some cities for access to better quality public schools (though measuring the quality of the schools :

http://cedb.asce.org/cgi/WWWdisplay.cgi?0609543
http://ideas.repec.org/a/jre/issued/v18n31999p3...
http://www.econ.yale.edu/seminars/labor/lap04/s...

People may or may not be short-sighted about their retirement (I spent my teens and early twenties studying and travelling, rather than saving for my retirement, and I don't regret it), but many parents are willing to spend on their children's education.

As for the dowry, I don't understand what you are trying to say. If no parent cares about their children's financial future for that child's sake, then why would the culture think it very bad to not provide a dowry? Especially in a culture that devalues women relative to men? And I am well aware that the dowry comes from the bride's parents, that is why I was talking about daughters in the comment you responded to.

Re: the dowry example – I suspect that's as much about fear of loss as gain for your children. It looks very bad to not provide a dowry. Further, those cultures generally devalue women relative to men, so the dowry is a form of compensation, too, since the dowry comes from the bride's parents, not the son's.

Well one possibility is that you perform a big study trying a variety of different interventions, and including control groups, and you see which interventions perform the best. That's what Project Followthrough did. And what the results from that imply is that teacher quality is not independent of school quality.

Having a massive variety of variables that affect outcomes is an extremely common occurrence in the real world. Take building a wind turbine (if you have moral objections to wind turbines, please substitute another engineering project of your choice). The materials you need in your wind turbine need to be both strong enough to stop the wind turbine from falling apart in high winds, and light enough that the wind turbine can operate at much lower wind speeds. You need to chose a site for wind turbines with good wind speeds, and also one that you can get access to install the wind turbines on a suitably strong base. The gearing inside the turbine in the engine needs to control the frequency of the electricity despite variable wind speeds. You need the tranmission connections from the turbines to wherever the power goes, which come with their own host of issues. Once you have installed the wind turbines you need a sufficient maintenance schedule to keep them operating. How do you choose the mix? Well, manufacturers pull this off all the time.

Another example is medicine. While of course individual patients have managed to survive actively-bad medical care, overall outcomes are improving by the healthcare system getting a lot of things right. If a patient is wrongly diagnosed, then that's generally bad for the care (this happened to me). If the patient is rightly diagnosed but the doctor prescribes the wrong drug, then that's generally bad. If the right drug is prescribed but the pharmacist dispenses the wrong drug, then that's generally bad (this happened to my Dad). If the patient picks up an infection while at the hospital then that's generally bad. The healthcare system needs to get a lot of things right. Often of course they don't, but it's hard to imagine that they could improve outcomes just by going for one factor.

If we are to have effective schools we can't just focus on one thing, we have to find a mix that works. Which we have. Now we face the problems of getting schools and school districts to implement it.

so the fact that low-income kids did better under one curriculum doesn't say much without knowing more about those kids.

Why not? This was a large project. From the link I supplied in my previous comment, 9,255 students were evaluated in the schools that received interventions, and 6,485 in the control schools.

Finally, I don't doubt that educational reform can help students, but lets not oversell it.

What are you talking about? What have I said that was overselling?

I am really surprised. I point out the example of an educational reform that has shown to be effective at improving the educational outcomes of deprived children. You don't say “Oh, that's great!” or “Oh, that sounds positive, but I have a few questions.” Instead you worry about overselling? What am I missing here?

Also, as I stated before, I think SES is overrated in explaining educational outcomes, so the fact that low-income kids did better under one curriculum doesn't say much without knowing more about those kids.

Finally, I don't doubt that educational reform can help students, but lets not oversell it.

Teaching should be open to anyone with a pulse and a college degree—and teachers should be judged after they have started their jobs, not before.

This scares me a bit. After all, teaching is not like sports. The teachers are teaching real students, and if those students have an incompetent teaching their education really suffers. If the incompetent teacher happens to be the very first teacher the kids get, so the kids are not properly taught to read, then the kids' education is going to be harmed for the rest of their careers. Even if the kids eventually run into a teacher who can correct their reading problems, there will still at least have been a year lost of learning vocabulary and practising reading skills, and in all the other subjects that depend on reading and a year is a big issue for a five-year old. Furthermore, bad teaching creates problems for all the future teachers, who then having to cope with a variety of levels of educational quality.

There will always a learning curve in teaching, as in any other skill, but on the other hand students do need some protection. Would you go to a doctor whose competence was not judged before they were let loose on patients for the first time? Would you drive a car which was only judged based on performance in the real world, and not on a testing track first?

Trainee teachers should have opportunities for real-world practice, properly supervised. But we should be assessing their skills before they start their jobs, at least to filter out the really bad ones.

Look at the fees charged by private schools some time.

And if it is taxation, I doubt even more the willingness of other individuals to sufficiently finance the education of other parents' kids.

I am interested in seeing if I can alleviate your doubts. After all, education spending strikes me as an incredibly popular political programme, one with strong support from the general public. But before I can do so, can you please define what you mean by “sufficiently finance”?

There are two problems with the words “sufficiently finance” without further definition. Firstly, there is far too much information out there in the world for any one person to learn it all given the maximum human lifespans so far recorded. We could spend all our spare resources on education (beyond those necessary to support human life), and yet everyone would still have something more to learn. But I do not think that is what you mean by “sufficiently finance” as that would imply no spending on health care or sewage systems or water quality or scientific research or civil defence or art or tasty food or warm houses.

Secondly, “sufficiently finance” assumes that if we spend more money we get more education out the other end. However, the education sector is notorious for the lack of correlation between spending and outcomes. See for example http://www.ft.com/cms/s/0/b5e0da2e-65c2-11dc-9f... . I believe that the main problem with education policy is how to improve the effectiveness of existing spending. However, people in the education sector who would benefit from extra spending have a tendency to always claim that more money would improve matters, without ever stating what this quantity of money is.

So I would feel much more confident of alleviating your doubts if I knew what level of spending you think would “sufficiently finance” kids' education.

This is wrong. Better educated people on average earn more money than non-educated people. (There are many exceptions, of course, but this holds true as a general rule). In many cultures it is extremely common for parents to be supported in their old age by their children, including in Western cultures until relatively recently. Therefore parents did gain financially from their children's wealth. And even in Western culture nowadays I know many examples of children giving money to their parents. I also know many cases of children paying for medical care for their parents, while this may not be an example of a weath gain strictly speaking, I do think that the prospect of better health care provides an incentive for better education.

And note that this analysis assumes that parents only care about their own wealth. However there is ample evidence that parents care about the wealth and happiness of their children in and of itself, even if the parent never gets any financial benefit directly. For example, in those cultures where a dowry is a common marriage tradition, why would any parent provide a dowry unless they cared about their daughters' happiness and future prosperity?

There was a big education study called Project Followthrough, which looked at whether low-income kids could be taught better by trying a variety of different school reforms. It found one curriculum that was successful – Direct Instruction, was successful in bringing low-income kids' educational performance up to the performance of middle school kids.
See http://www.projectpro.com/ICR/Research/DI/Summa...

The Direct Instruction curriculum consisted of the following main features:
– A set of lessons in reading and mathematics, written and field tested to provide unambiguous, clear and conscise explanations of all the skills necessary for reading and basic maths.
– The set of lessons also provided many opportunities for the teacher to seek feedback on how the kids were understanding the lesson (the general rule was that the students should be providing a response ten to fifteen times per minute), and to reshape the lesson based on that feedback.
– the high rate of student responses also meant that the kids had many opportunities to practice the skilsl they learnt.
– Kids starting at a Direct Instruction school would be tested on their existing knowledge in maths and reading, and be placed in the lesson sequence based on their performance, so a kid who already knew how to write their name would be placed later in the reading sequence than a kid who started school not knowing their alphabet. A kid could and often was placed in different lessons in reading and maths.
– Every teacher at the school taught reading and mathematics at the same time. Kids were divided into groups of 5 to 7 students all at the same point in the lesson sequence. The teacher would spend the session working with these small groups in turn, while an aide supervised the remainder who were doing more independent work. In one school there wasn't money for the aides so every adult at the school was roped in for this bit with reading.
– Kids' placement in the sequence was often reviewed, so the kid who started not knowing the alphabet could wind up surpassing the kid who started knowing how to write their name, if they happened to be a faster learner.

Some sample DI lessons are available at http://www.specialconnections.ku.edu/cgi-bin/cg...

Teacher quality is affected by far more than the individual teacher. There's also the support that surrounds those teachers.

I do not deny the importance of teacher quality. There are super-teachers out there who can succeed with shoddy curriculae, no textbooks and despite an unsupportive administration. And there is the odd teacher who should not be anywhere near any children, no matter how good the curriculum. But, when it comes to educational reform, Project Followthrough shows that we shouldn't just be thinking about teacher quality vs student quality, school quality is at least as important.

I agree with what you're saying, but I don't think that it is reason against making the market for teachers more competitive. Saying we can't show a correlation between student performance and teacher quality with todays population of teachers isn't indicative that we also can't with a more competitive system of hiring teachers.

Is teacher quality irrelevant? No. Will increasing teacher quality dramatically help the students who need it the most? My prediction is no. I think that student quality is both (1) the main predictor of student outcomes and (2) that student quality is fairly non-malleable.