Tuesday, December 20, 2011

Does Jason Glass know better than you do?

The latest “education reform” talking point is that class size doesn’t matter. It’s been empirically proven!
[Economist Roland] Fryer found that class size, per-pupil spending, and the number of teachers with certifications or advanced degrees had nothing to do with student test scores in language and math.
. . .

Schools that focused on teacher development, data-driven instruction, creating a culture focused on student achievement, and setting high academic expectations consistently fared better.
I remember taking an SAT prep course when I was in high school. We all listened as the instructor worked through exercises designed to get our test scores up. I suspect it was all very data-driven. If that’s your vision of education, it’s probably true that class size is largely irrelevant. And who could want more for their children?

One of the people arguing that class size is unimportant is our State Director of Education, Jason Glass. “Given where class sizes currently are in most schools in the United States, I am willing to trade holding the line or even slightly increasing class size in exchange for improving educator effectiveness.” With that stance, Glass couldn’t get elected to any school board in the state of Iowa. Yet he has more control over our kids’ education than any school board has. What’s wrong with this picture?


Anonymous said...

Hi Chris - good thing I'm not elected!

We've been chasing the tail of lower class size in the United States for nearly 30 years, but our results haven't improved. While there is some evidence to suggest that very low class sizes (below 15) matter in lower grades, the macro level evidence doesn't seem to show an impact.

While I'm certainly not advocating for DRAMATICALLY trying to INCREASE class size, I do think we can use the limited resources we have in much more strategic ways than the relentless pursuit of lower class size.

Nearly all the systems internationally that are outperforming us place more emphasis on educator quality than class size. In your own back yard of Iowa City (and one of my favorite places!) you have a school system that is consistently high performing, yet has some of the highest class sizes in the state.

Class size matters, but I don't accept that it's the panacea some believe it is. My point is that it has to be considered against the other options you have for using the finite resources available.

I'd encourage you to review the recent Brookings Institute report on this very issue, that I think lays out the arguments on both sides nicely.


Thanks for raising the issue - I enjoy your blog.

Jason Glass

Chris said...

Jason -- Thanks for commenting! But tell me: Do you really think parents want smaller class sizes because they think it will raise standardized test scores?

Mandy said...

Jason, you state that ICCSD has some of the largest class sizes in the state, yet consistently outperforms other districts. Could it be that there are other things that might contribute? Are the demographics the same at these schools? What about rates of free and reduced lunch at the schools? What about the percentage of children with special needs?
Lets look closely at an example. One 3rd student is in a math class with a class size of 3 students and one teacher. Looking at his test scores, he is performing at the bottom 1 percent of 3rd grade students. Having a small class size doesn't seem, on the surface to have an impact. Maybe if we increased teacher quality and didn't focus on class size we would be using our resources much more effectively. Maybe if we scratch the surface a bit more there might be other causes? Maybe the teacher is fantastic and creative and experienced and the student has made tremendous growth from 2nd grade to 3rd grade but the grade level test doesn't show this growth, only that he's at the bottom 1 percent of 3rd graders. Maybe this kid comes from poverty? Maybe there are other factors at play? How can one say we just need to look at the data to prove the point the smaller class size doesn't have a huge impact and we're only raising the size a little bit yet at the same time one isn't even looking at ALL of the data?