Charters etc up, reading comprehension level, null hypothesis?

Pro-charter school group estimates 14 percent enrollment gain nationwide - The Washington Post

The National Alliance of Public Charter Schools estimates in a new report that 2.9 million children now attend U.S. charter schools, up 14 percent from last school year.

Note that 14% increase per year corresponds to a doubling time of just under 6 years...quadruple in twelve...eight times as many in eighteen years...well, maybe not. I do think the proliferation of charters is a Somewhat Good Thing, partly because choice is a way to get people to be happy (even if actual results are no better, I think happiness is a good thing and if you don't, well....okay.) Mostly, though, because I like experimentation; each public charter (or in some cases groups of charters) is an experiment in doing things a bit differently than whatever failed in the area it's trying to serve. We don't know what works for which students in which kinds of learning, and we won't find out without trying, and that will involve a bunch of failed experiments and failed programs and failed schools, which is sad but necessary. (500 charters opened and 200 closed.) Here's Grant Wiggins, thinking about public schooling mainly, on the way that zillions of trials (and hugely rising expenditures) have utterly failed to improve average reading comprehension: Maybe we don’t understand what readers really do – and why it matters | Granted, and...

Numerous causes and their implied solutions, as readers know, have been proposed for flat reading scores: poverty, low expectations, inadequate background knowledge, an anti-boy bias in schools (especially in terms of book selection), IQ links to reading ability, computer games, TV, etc. etc.

The utterly flat national trend line, over decades, says to me that none of these theories holds up well, no matter how plausible each may seem to its proponents. Perhaps it’s time to explore a more radical but common sense notion: maybe we don’t yet understand reading comprehension and how it develops over time.

If we don't specifically know what it is we're hoping they learn, we don't know where they're going and it's hard to help them get there, even if we have tests that supposedly tell us whether or not they've arrived. Of course maybe some or most or all of the efforts really are working, but they're roughly balanced by negative effects: maybe the harder we try the more they fail. As Wiggins put it last month in A Post from Paris | Granted, and...

A 7th grade girl, when interviewed by teachers as part of our summer institutes, said the most amazing thing when asked how she felt about ‘typical’ teaching. “The more the teacher talks, the more I feel alone and useless.”

Or maybe the truth is close to Arnold Kling's "Null Hypothesis". The Null Hypothesis in Education is Hard to Disprove | askblog

In education, the null hypothesis is that nothing makes a long-term, scalable, replicable difference. That is:

1. Take any pedagogical innovation or educational intervention.

2. Subject it to a controlled experiment.

3. Evaluate the experiment’s outcome several years later.

4. If the experiment works, attempt to replicate the experiment in more situations.

By the time you reach step 4, if not sooner, you will be unable to show that the innovation makes any difference in outcomes. What this suggests to me is that in the long run it is the characteristics of the students that determine outcomes, at least on average. Think of an individual student as “predestined” to reach a certain outcome....

That's an extremely uncomfortable way to think, even at the margin. It's obviously false in extreme cases, such as urban poverty, but it's not obviously false for the programs at the margin, i.e. programs that are supported and opposed with reasonably thoughtful people on both sides. Kling notes an OECD report saying The Null Hypothesis Strikes Again | askblog

In practice, however, grade repetition has not shown clear benefits for the students who were held back or for school systems as a whole.

One interpretation of this is that the marginal benefit of an additional year of schooling is zero. However, that interpretation is not something that anyone wants to discuss.

I don't actually believe even the marginal version of the null hypothesis...I think that a sufficiently-responsive educational system could actually evaluate each kid and in almost all cases say "this is the group they fit in (diagnosis) and this is what worked on members of that group (i.e. what they should do next)." In software. On a cell phone.

But not this year. :-)