April 3, 2008

Slate on NCLB Contd

(continued from part I)

Now we get to part of the article in which Ryan tells us how to fix NCLB, otherwise known as pure fantasy. He has three main ideas for "fixing this mess," and by ideas I mean Ryan's opinion utterly lacking any empirical support.

1. National Standards.

It's time to create national standards and tests in at least reading, math, science, and social studies/history.
For some reason, national standards have become synonymous with "perfect standards." As if anything that goes through the political process on the national level with all the competing interest groups get a chance to influence the process in a way that benefits their particular way of educating. Now recall that most educator's "particular way of educating" is not only ineffective, but often detrimental to at-risk students. To put it mildly, there is absolutely no reason to believe that National Standards will be any better than the existing state standards, which are on average pretty awful.

2. Administer fewer tests

National tests should be given less often, perhaps in only fourth, eighth, and 11th grades.
Because it's always easier to identify and diagnose problems with less feedback, especially when you don't know what you're doing in the first place. It's a trial and error approach in which each trial last four years before we determine whether there's been an error.

3. Rank schools; don't prescribe punishments

Ranking systems aren't perfect, but using multiple criteria to rank schools should provide a much clearer and fuller picture of school quality. States can then decide on their own how they want to sanction or assist the low-performing

I don't see how ranking schools is any different of better than what we're doing now, unless there's some benefit to even less accountability--because that's exactly what everything proposed in this paragraph actually leads to.

And, before NCLB we used to let the states police themselves and they completely failed to do so. So good luck with that.

Then Ryan tells us:

If and when NCLB is fixed, the next president should concentrate on two key
issues: teachers and preschool.

By concentrate, Ryan means "give more money to." And, we all know how well that's worked in the past. There is little correlation between teacher compensation and teacher performance. Most of the existing problems are unrelated to compensation. Increasing compensation won't help poor teaching practices and bad curricular decisions.

Universal preschool is the next magical panacea. But I can't for the life of me figure out hoe adding a year of preschool is going to help at-risk kids if the same clowns that run kindergarten run it. If they can't teach them in kindergarten, which they can't, what makes you think they'll be able to teach them in preschool when they are even younger and more difficult to teach.

NCLB is far from perfect; but there's no reason to think that Ryan's suggestions will improve anything.


Catherine Johnson said...

Well...I'm ready for mathematicians, historians, etc. to create their own set of standards & publicize the heck out of them.

The national history standards are great; some states adopted them. (New York state has them.)

And who's drawing up the document of standards for Algebra 2?

Isn't that Achieve?

National standards drawn up by the federal government and them imposed on all 50 states are going to be a disaster.

Michael Shirley said...

By concentrate, Ryan means "give more money to."

You're assuming this is true; it's not in evidence.

KDeRosa said...

From the article:

"To attract more and better qualified teachers, and to attract them to particular subjects and particular schools, we need to pay them more."

KDeRosa said...

The defense rests.

Michael Shirley said...

Ken, you've said that "concentrate," as Ryan uses it, means "give more money to." As you've put it, it only means that. Your quotation does not support what you've said, and the jury (okay, the prosecutor) finds for the prosecution. Care to appeal?

Brett Pawlowski said...

Ranking systems are actually a really good idea, as long as you're ranking something that matters, like value-added performance, and as long as it's coupled with an assessment of grade-level proficiency. It's like checking your driving status by looking at both your speedometer and at the mile markers.

I've been working on a school ranking initiative in Tennessee using their value-added data, and it provides great information - information that the public has been very responsive to.

Where Ryan is completely wrong is in thinking that multiple measures can be used to clarify, rather than obscure, the picture.

When we published our rankings of value-added performance, the district countered by saying that value-added is only one of many ways to evaluate, and that other factors, including graduation rates, promotion rates, attendance, and several others should all be factored in to a single composite measure. (ie, all our schools are equally great, they're just great in different ways.)

First, such a system combines apples and oranges: a system that actually tells you something about school performance (ie, value added), and several measures that not only don't speak to academics, but that can be manipulated by the schools and districts. Promotion, for example, is determined by district policy, not academic achievement, and in fact including it as part of a composite ranking provides them a disincentive for enforcing academic rigor: a lower promotion rate, while likely better for student learning, would look bad for the district.

So, yes to rankings, as long as they measure something relevant; no to multiple measures.

KDeRosa said...

Michael, i don't think that is a fair nor the only reading of what I wrote.

Tracy W said...

a lower promotion rate, while likely better for student learning

I'm doubtful about this. Surely the best thing for student learning is to teach them well the first time around?