Public Education, in the United States, is a topic that is fraught with a very large number of problems for productive conversation. This is the first in a series of posts in which I will attempt to move this conversation… somewhere productive.
In order to have a workable conversation, it’s probably best to start by pointing out some of the problems that make this conversation (typically) non-productive. Since public education is entirely political, and since the United States has a very coarse filter for political affiliation, your initial biases have a tendency to process you immediately into the Big Sort of LEFT-RIGHT. For the record, here, I’m not talking about your tendency to vote Democrat or Republican, here.
I’m talking about how the default sort affects your *conversations*.
Once you’re initially sorted, the conversations you will be typically having, and the implicit assumptions they have, will go a long way towards making it impossible for you to have a constructive conversation with anyone who isn’t already sorted into your bucket. Default assumptions are wildly different. If you don’t believe me, talk to a fairly liberal public school supporter from Louisiana and a fairly liberal public school supporter from California. You’re going to be having two wildly different conversations.
So I ask – if you’re going to participate in this conversation – that you first consider your bucket, and your biases, and how they are going to decidedly skew your perceptions of what public education is about.
Because otherwise we’re not going to get anywhere.
The first, biggest mistake is that almost everyone has a misunderstanding how public education works in the U.S.
That is, that they think there is a quantifiable thing called “public education”.
Public education is decentralized. In the U.S., it’s decentralized into quite possibly the single most extreme example of our layered, federated system of government, more that any other public service provided in the United States. Your local water board has only a few couplings to your state water board, and there’s a handful of laws at the federal level that affect water rights (and by “a handful”, I admittedly mean “a crazy amount”, actually… but it’s a vanishingly small number in comparison to public education).
There are many more systems of “public education” than you think, and every one of them is stacked upon at least one other, with very little in the way of coherent governance models.
rO to put it another way… everybody is messing in everybody else’s business far too much, or too little, or both. Usually both, and often in diametrically opposed ways, come to think of it.
How does the governance structure work?
First, you have the LEA: the local educational agency. This is your local school board. (I’m intimately familiar with this problem space, as I serve on one. Reminder: standard disclaimer, my thoughts on this post are only my own and not necessarily part of the collective board.)
As of 2010-2011 there were 13,588 local school boards in the United States, according to the Department of Education.
For large states you have county boards of education, between the LEA and the State Board of Education. In California alone there are 58 county boards of education, with 334 elected board members and 7 appointed ones. Then you have state/region boards of education… one for each state (50), Washington D.C. (1), Puerto Rico (1), Guam (1), the U.S. Virgin Island (1), and American Samoa (1).
Each state has its own method of selecting school board members, state school board members, setting teacher license standards, adopting curriculum standards, etc.
The NASBE provides a handy (PDF) matrix of the states and regions. Prior to Common Core, almost all of those states had different standards, which means they were all testing different things at different levels of education. Even with Common Core, we’re not all that common.
Each one of those state boards interacts with county boards that in turn interacts with local boards. On top of that conversation, you have federal resources and obligations that come into play, many of which bypass the normal governance tree. The federal government puts one set of obligations on state boards, but other sorts of obligations directly upon local school boards, and there isn’t any organizational structural reason for this, it’s the (usually arbitrary, sometimes capricious) collective will of Congress that sets these rules.
Oh, and there’s the Bureau of Indian Education, and the DoD schools, so you’ve got those kids, too!
There’s no “public education in the United States”, in a meaningful analytical sense.
There are “public educations”, but they’re all different, often in very significant ways.
What does this mistake mean, for conversations about public education?
In social science research, you have to be very careful in your problem domain selection, because you have to make sure you’re studying things that are reasonably similar. It’s the first thing they teach you in your research methods class.
Well, if you are including (things) that aren’t (sufficiently like the other things) in the same analytical bucket, you don’t have a solid connection between independent and dependent variables. That means you can be trading off a huge amount of precision when you set too generalized parameters on your study space.
Note that this is true in hard sciences, too, obviously, it’s just a mistake that you’re less likely to make.
If you’re doing material science and you’re testing samples of materials, but your samples are biased towards most of the materials having a common condition (they are brittle, or flexible, or resistant to compression, or whatever)… and you draw some general conclusion about materials based upon average properties, you’re going to build a lot of terrible buildings, right? That’s pretty obvious.
And yet almost all thinkpieces about public education in the United States rely far too much, if not entirely, on general conclusions based upon average properties.
I’m not talking about research studies (although there is plenty of that), I’m talking about think tanks and the sorts of punditry that we pretend are real policy conversations.
Folks can (and do) make enormous category errors in their analysis whenever they’re studying public education effects.
As an information consumer, to prevent your own biases from running roughshod over your understanding of public education, you must be immediately wary – I’d argue to the point of supreme skepticism – of anyone that uses national statistics to say anything specific about public education in the United States.
In fact, if the thing you are reading is making a strong claim about the general case of public education in the U.S. it’s overwhelmingly likely to be basically just hot garbage cooking on a dumpster fire.
This goes for folks who try to compare U.S.-wide public education outcomes to other countries (Finland is a popular case here), folks who try to draw generalizable conclusions about how public education spending affects outcomes, or that argue that school closures work to produce better outcomes, just for a few examples. To the extent that they have points (and sometimes they have points), the points are tied to ideological or normative assumptions that have very little to do with general education policy. Certainly not workable, nation-wide education policy.
California has 1050 school boards, about 8% of the total school boards in the country. But we have over six million public school kids in California, almost 13% of all public school kids in the country. Texas has another five million. Combined, those two states represent one quarter of all of the public school kids in the country.
Well… that’s complicated! How do you define “less”?
(We’ll talk about that next time.)
But to stick to the topic at hand, in the broadest sense it’s true that California and Texas both spend less per pupil than most states. Can we draw any conclusions from this fact? Well, from that alone? From an education policy standpoint? Not really, not strong conclusions anyway. Look at the map! Texas and California both appear to be green and red, depending upon county.
Of course, the number of kids aren’t distributed evenly around the state, either. Most of the kids… are in the red zones. But we’ll get more into depth on that next time.
In the general case, combined, those two states are both outliers on total population, and outliers on per-pupil funding.
From an analysis standpoint, we can therefore draw one conclusion:
If you’re talking about “public education in the U.S.” and you’re drawing conclusion based upon country-wide results, you’re probably doing very bad analysis when 25% of the kids are all in two states, and they are funded less than most. Your cost-outcome analysis is going to be irrevocably skewed by those two outlier facts.
So if you’re reading something that comes to a conclusion (particularly one you already agree with) but it relies on this sort of analysis… stop.
Don’t do that.