Here are hypothetical questions based on situations I have witnessed: One teacher is a grandmotherly type who is “relaxed” in her teaching but beloved by the children. Her students love coming to school each day. Another teacher in the same grade is very focused and has well-planned lessons each day and prepares her students carefully for the tests at the end of the year. Her students do much better on the tests than “Grandma’s,” but they can’t wait for school to end each day. Who is more effective? How important are the children’s attitudes toward school? Can they be accurately measured? Are they more important than test scores? The variables are just endless. I am not a scientist but I know that I don’t know. Remember that a lot of the research in education is done so an academic can earn tenure; that’s the purpose of it. Yes, we should continue to seek answers to these questions, but let’s proceed with a healthy dose of skepticism.

Too many of our kids come into the school systems without the social skills, character traits and values that allow them to learn successfully in a classroom enviornment. This has a negative impact on EVERY single aspect of the learning environment.

It doesn’t matter if you have a classroom size of 20/25/30 — even a few kids who are unruly and disruptive, who lack respect and consideration for the needs of others, can ruin the learning experience for ALL.

The reality is, we will NEVER be able to reduce class sizes to the point of educational nirvana. Smaller classrooms means more classrooms, means more teachers, means more schools, means more administrators, ad nauseum.

On the other hand, if, as statisitics show, our teachers today are losing 30/40/50% of productive classroom time dealing with disruptive students, this IS something we have a chance of dealing with.

It’s not about reducing class sizes to the point where they are “manageable” — it’s about managing the classrooms sizes we have and with an emphasis on social skills education and discipline in the schools, we can improve the learning opportunities for everyone involved.

And I think the main reason why researchers in medicine, economics and so forth laugh at us is not our committment to statistical findings, but the quality of the data that makes up those findings. I’ve seen that data as it’s created, and frankly, most of it is garbage.

Many of the mistakes come frome reading just the Executive Summaries and not considering context, as Tom also reminded us by citing Hanushek. California, for instance, ran into that problem. If you don’t have enough of a supply of qualified teachers, then class reduction efforts can backfire.

So, the question should be how we recruit and retain more teaching talent. Reducing class size is one way. Protecting teachers’ due process and contractual rights is another essential ingredient. We can reform seniority and evaluations, but this loose talk about laying off teachers by perceived effectiveness, as opposed to negotiated agreements, would be a horrible mistake.

Proposals that would turn teachers into widgets, even if the report that recommends it is entitled “The Widget Effect” would just make things worse. I’ve have up to 70 seniors for a month with few problems though they learned less. Last year my class with 44 sophomores never recovered from the overload. I suspect that the tipping point in a tough class of freshmen ordinarily is in the mid-20s. But add just a few more students who are mentally ill, Seriously Emotionally Disturbed, who have just buried a family member, or a gang war, and the tipping point drops dramatically.

If you believe that statisical formulas can approximate these realities, I’ve got a research project for you.

You do realize that researchers have methodological techniques that enable them to disentangle two scary independent variables. It’s called multivariate regression and has only been around for decades. Therefore, your analogy with the lawn chairs just doesn’t really apply. The research that has looked into class size finds that its independent (in other words holding all other things constant like teacher effectiveness) impact on student outcomes is minimal (the major exception as Andy noted being in lower grades and with huge cuts in size, i.e., The Tennessee Study). The research also shows that the independent effect of effective teaching carries tremendous weight and is the single largest school level factor capable of increasing student achievement from its baseline level.

I mean that’s just what the research says. It said it yesterday, it says it today, and it will say it tomorrow. No amount of personal experience is going to change that. It’s called peer reviewed for a reason: it observes a huge N in a tightly specified model and replicates it over and over again so that we don’t have to rely on “individual” experiences.

Would anyone take a drug that the American Journal of Medicine did a huge study on and found it killed people or didn’t work just because they personally thought it gave them some results? Pop away!

Of course teacher effectiveness is more important than class size. If you put a folding chair in front of seven kids and a live teacher, effective or otherwise, in front of 27 kids, the kids with the teacher would learn more than the kids with the chair.

But that’s not the point. You’ve got two independent variables: teacher effectiveness and class size. They both have a profound effect on student learning. We don’t need a study to tell us that. What we need is a way to get better teachers in front of smaller classes. Policy makers love studies that get them off the hook. If we got the average class size down to 16 students, the researchers would be amazed at the sudden increase in teacher effectiveness and student learning.

If the research says small classes don’t make a difference, I want to know what they consider “small” and where the class sizes were before they got “small”.

And my MA didn’t make me a better classroom teacher. I’ll support that research.

Love the blog and the discussions.