Wednesday, June 2, 2010

Statistical Sleights of Hand

How can the government cut the number of teachers while at the same time promising to improve educational results? By claiming that class size does not significantly affect outcomes, which is what Luc Chatel has just done. Here, econometrician Arthur Charpentier explains what the minister left out of his convenient analysis. And here, a discussion of what is at stake.

And here is a discussion of the latest DARES statistics concerning les heures sup.

2 comments:

TexExile said...

Well, yes but...

1. The econometric analysis to which you link your post is indeed hokey and simplistic beyond belief.

A close colleague who has worked on these issues for years, on a number of cross-country studies, tells me that, in a nutshell the evidence shows that changing class sizes on average probably doesn’t matter that much.

HOWEVER, it can have quite complex effects depending on what it does to average teacher quality, the impact of and on different teaching techniques, and how the change in class size is distributed. (For example, it can make a big difference to struggling, disadvantaged kids and little difference to well-off kids with lots of social capital from their parents.)

2. The current situation in France is very different from the US – the threatened teacher layoffs will return class sizes to about where they were 10 years ago – before a big recruiting binge. In contrast, French class sizes have remained around constant. That is not an argument for or against what the government is trying to do but it is important not to confuse the two debates.

MYOS said...

So, essentially, my kid not only may have a teacher with zero training but he'll have even more classmates?
In the US, classes can have anywhere from 12 to 25 kids - depends on the subjects, but usually the more advanced the fewer kids. Also, in many States, there's a cap at 18 or 20 kids up to 3rd grade.
Right now, I'm talking about a primary class with 27 students and a middle school class with 29.
Based on the document, the average addition would be one student (in order to cut 10,000 classes/classrooms), but if you look carefully you'll see that they plan to add 4-5 kids wherever numbers aren't maximized. In my opinion, 20 elementary kids and 25 high school students would sound like it's "maximized", but apparently not here. Since the maximum number of desks in those classrooms seems to be 30 or 32, I bet they plan on having 32 kids per room everywhere.
I can't do anything about it.
And I CAN'T homeschool.
So, I'm quite frustrated with French schools right now (and no, there aren't American schools around).