And what about in real life? Subsequent to STAR many districts, at great expense, rushed to reduce class size. Where’s the predicted stellar results?

And minimal? There were significant gains

Effect sizes were less than a quarter standard deviation (in most cases much less) which is not educationally significant. Such a small effect would only boost a 20th percentile school all the way up to the 28th percentile. Break out the champagne!

if memory serves me right, grades 1 – 3, that persisted, measurably, all the way through 12th grade.

Which proves my point.

The gains, such that they were, took place mostly in K, not in grades 1-3. Then the kids were dumped back into regular sized classroom in grade 4.

At best, STAR tells us to reduce class size in K and then only for the low-SES group since the other groups’ effect sizes were miniscule. Apparently, keeping existing class sizes in classes thereafter seems to be acceptable to hold onto those gains. At least there is no evidence yet to suggest otherwise yet.

This doesn’t seem to be the outcome you were suggesting.

No one else has conducted a study of that magnitude or duration. STAR may be the only ed study with a large enough sample and a strong enough design to get results that show above the noise.

And minimal? There were significant gains in, if memory serves me right, grades 1 – 3, that persisted, measurably, all the way through 12th grade.

Anyway, that’s enough. I am tired of reading here what I could read at the Hoover Institute’s site, but choose not to.

Here’s what I do know. When I was a NYC student, class size was typically in the 30-34 range in middle and high school, a bit lower in elementary school, not all that different than what it is now. My oldest was in class sizes in the mid 20′s years ago in the burbs. That hasn’t changed much either. To say class size has plummeted is a nice sound bite but I’ve seen no evidence of plummeting class size over the past 30 years. Somewhat less? Perhaps. Plummet? Come on

There’s also no evidence that teachers are being paid drastically larger amounts of money in the NYC school system. Even if we were count the raises of the last two contracts as real raises (they weren’t when you factor in the longer day) the average teacher salary in NYC has not kept pace with inflation over the past 30 years. Like everything else pretty much, things go up in price, but I hardly see where my current salary covers the increases in housing, gas, tolls, insurance, etc. in the region. That’s not to say private industry has it any better. But NYC would have to jump teacher salaries another $10-$15,000 a year to noticably make a difference that would attract a larger number of applicants with superior teaching credentials. And that’s optimistic.

Finally, your SES stats lead me to only one conclusion and it’s not one I’m prepared to make. That is that children from low income families are going to perform poorly no matter what resources are thrown at them. So why waste the money. That’s an attitude I can’t accept. I’m realistic. NYC students are operating at a deficiency in terms of family consistency, primary language learned at home, and parental involvement in the classroom. But just because we might not achieve substantial test score increases doesn’t mean we should opt not to “waste” money on them.

I still believe class size isn’t a panacea but rather one weapon in a school system’s arsenal to manipulate student performance. I still think that any system that doesn’t have as its fundamental goal, a higher percentage of teacher candidates coming from stronger backgrounds and retained for a long enough period of time to become useful, is a system destined to fail.

I’ve read it.

You may have read it, but you clearly haven’t understood what you’ve read.

I HAVE read the study and it does not say what you think it says. I’ve also read many articles both pro and con on the STAR findings. I’ve also seen no one since the project concluded succesfully get the same results, minimal though they were, that they did in STAR.

Again, you’ve failed to answer the issues I’ve raised. Sounds like you’re a teacher (though clearly not a teacher of statistics or the scientific method), why don’t you try to teach me what your position is instead of relying on cheap shots.

Why don’t you tell us what you think STAR supposedly stands for?

This is embarassing. You are attacking work you haven’t read. That may go over fine in dinner conversation, unless someone in the room has read the stuff.

I’ve read it.

” … or can’t you think of a real response to to my critiques?”

The critiques do not appear to be yours.

Bad instruction is bad instruction.

And bad instruction is what you have in NYC with only 1/3 of students at levels 3 and 4. Whatever is going on in the schools it isn’t working for 2/3 of kids.

It multiplies exponentially in a large classroom for two reasons.

Ok, so let’s pretend we can get the same effect size (.25 SD) that STAR got. We’ll cut class sizes down to 13-17. In your typical inner city school peforming at the 20th percentile, we’ll boost performance to the 28th percentile with a .25 SD increase. Now insteads of 80 students not making the grade, only 72 won’t under the paulrubin plan. It’s time to celebrate. Where should we erect the statue in your honor, assuming the city has any money left now that we have to pay for twice the number of teachers?

#1, if class size isn’t that important, why not increase it and save money

Seems to work OK in the asian countries so maybe you have a point.

if class size reductions would be beneficial to certain groups of students not necessarily classified as special ed, why isn’t the all caring, all knowing NYC DOE striving to achieve that instead of playing games with curriculums and bulletin boards and high cost items like extended school days for subsets of the population.

One, possibility is that they don’t know what they’re doing. Two, because the expected effect size is small and speculative and the expense would be enormous. That’s lose-lose.

Class size needs to be flexible. Optimally it should consist of a combination of large group lecture and small group instruction to get the best of all worlds.

Maybe this’ll work, maybe it won’t. Surely you have evidence to back up your contention since you speak with such authority. Show me the inner city school where this plan works and it’s test scores. Talk is cheap.

The teachers in the trenches need to be consulted and their opinions need to be heeded because they’re the ones who will do the implementation.

Yeah, I’d like to round up all the teachers who work in a typical inner city school who’ve managed to get their student performance up to the 50th percentile and see what they say. there’s probably 5 in the whole city.

We don’t need to hear from the rest.

If large classes are antithetical to profound learning at our level, what makes anyone believe that packing in children with various needs is going to be effective?

Trying to teach any size class full of students with various needs is pretty stupid in the first place.

Show me private prep schools or suburban schools that have overcrowded classrooms and parents willing to pay high tuition and/or school taxes and I’ll reconsider my position.

The one good thing about NCLB is that we now have disaggregated data for lower performing minority groups and low SES students. And wouldn’t you know it, once you compensate for SES, these subgroups don’t do much better in the affluent schools with small classes and lots of money.

That’s because what gets taught in the suburbs is just as bad as it is in the city. The differences are superficial.

If I can give my kids more attention, I’m going to get better results.

This is what every teacher says. Classroom sizes have plummeted around the country in the last 30 years while funding has more than doubled in constant dollars. Student achievement remains stagnant during this same period across the board. Explain.

Apparently, it’s not quite as simple as you make it out to be.

So basically you present this nonsensical position but refuse to respond to what I said. #1, if class size isn’t that important, why not increase it and save money (and reduce the number of weak teachers if handled properly). And #2, if class size reductions would be beneficial to certain groups of students not necessarily classified as special ed, why isn’t the all caring, all knowing NYC DOE striving to achieve that instead of playing games with curriculums and bulletin boards and high cost items like extended school days for subsets of the population.

This is still an absurd argument though. Class size needs to be flexible. Optimally it should consist of a combination of large group lecture and small group instruction to get the best of all worlds. So in a typical middle school math teacher’s program, there should be a period where all the teacher’s classes meet simultaneously in a small auditorium like classroom, preferably with ready access to technology like SmartBoards and the specialized software. The four periods saved from doing that twice a week lets you reduce class size for the small group instruction periods by close to 40%.

It’s time to think outside the box but one thing is an absolute. The teachers in the trenches need to be consulted and their opinions need to be heeded because they’re the ones who will do the implementation.

Are you stuck on stupid or can’t you think of a real response to to my critiques?