Monday, September 1, 2014

How To Spot Lies

It is easy to get caught up in the wrong conversation. If someone locks you in a room without food and tells you, "I'm doing this to make you a better person," it's easy to get caught up in a whole argument about how this will actually hurt you and why would going without food make you a better person, anyway?

But let's not have that conversation. Instead, let's reverse engineer the premise. You've said that the objective is to make me a better person? Let's start from there and ask this question:

If we were going to try to make me a better person, what would we do?

It's this same process that so often leads me to conclude that many reformsters are simply lying (or, at best, confused). They present a misguided, distracting or destructive policy and then at the end, tag it with a noble or worthy rationalization, like stapling a lion's tail onto the butt of an ugly donkey. What happens if we just look at the tail and ask, "What animal would this have come from?" Let's look at some reformster classics. that students aren't the victims of their zip codes.

This is used to justify charter programs, particularly large clusterfinagles like Newark.

But if we wanted to make sure that no child went to a lousy school just because she grew up in a poor community, we would make sure that no schools were terrible, regardless of the neighborhood. We would demand that state and local governments found ways to fully fund each and every school. We would make sure that no zip code anywhere was victimizing any children. That way there would be nothing that anybody needed to escape. that we know how well our children are progressing in their education.

If we wanted to have a really effective measure of student learning, we would have a long and difficult task to undertake.

First. we'd have to come up with an effective measure of all learning, which would be hugely challenging on a large scale. We would have to be relentless in making sure that we were designing instruments that measured what we wanted to know, and not basing them on what we can most easily measure. I'm not sure exactly what it would look like, but I expect that 1) there wouldn't be a multiple choice question anywhere in it and 2) it wouldn't be economical to do it on a national scale.

Then we would have to decide how to use the data that we gathered. We'd need long conversations about where to put the various dividing lines (cut scores. etc) and how to package the data in a way that was meaningful and useful to teachers and parents.

And we would have to decide whether our goal was to provide teachers with information to inform their teaching, students and parents to make their own choices and decisions, or suitable for government functionaries to make state and national policy decisions. Instruments that did one of these would be hard; instruments that did all three might be impossible. The easiest approach is the one we're currently using--instruments that do none of these things. that all students are taught by a great teacher.

If we want to do this (and why wouldn't we), there are several problems that have to be tackled.

First, we would have to identify the best teachers. This would require multiple instruments and some broad judgment. We would have to test, pilot, check, test again. And if experts in the field of these sorts of measurements said, "Hey, this thing you've come up with is crap," we would not ignore them.

We would have to recognize that all teachers start somewhere, so we would want to have considerable training and support through the first several years in the classroom. We would also recognize that most teachers hit tough patches now and then, and at those times they need support, not condemnation and threats.

And we would want to come up with ways to attract and retain the best people. And, because we know that stability in school is important, we would want to hold onto good teachers for their entire careers.

To that end, we would offer job security, solid career pay, autonomy, resources, and support to do their job. We would foster school atmospheres that treated teachers like managers, not flunkies. We would treat them like valued professionals, experts in their field, whose knowledge and insights would be a valued element of how the school functioned.

You can play this game with pretty much any reformster proposal (it works in the rest of life as well). If they say, "By doing this, we can get to X" just ask yourself, "If I wanted to get to X, what would be the best way?" Of course you're answer isn't the only one. This approach can be used badly (beware anyone who says sentences that start with "If you really loved me..."). But if you find your ideas about getting to X are wildly different from what you're being sold, something is up.


  1. I especially like the part about the zip codes. As far as seeing if our children are progressing, first we need to define progress. To make sure all teachers are good (I think it's best to start with shooting for all good teachers rather than great; good is pretty good, to me it means much more than adequate) we need to figure out the characteristics of a good teacher and then figure out what preparation will result in this goal. And we have to figure all this out before we can think how to measure any of it. Otherwise we don't even know what we're trying to measure and that's certainly putting the cart before the horse.

  2. Speaking of Newark, Booker joined a lot of other anti-public education people to attack Florida's voucher lawsuit, I imagine he says victims of zip codes all the time.