All things excellent are as difficult as they are rare.


12 April 2014

Freedom, Civic Responsibility, and Testing Opt-Outs

Chester E. Finn is upset.  Why is he upset?  Well, he's upset because apparently the number of parents choosing to exempt their students from state testing is on the rise.
The opt-out-of-state-testing movement has notched more wins lately. “Thousands,” we read, are refusing to take the tests in New York alone. And tons more interest and attention are being devoted to this topic in states and communities far and wide.
 I came upon his piece (as I come upon so much in the Edu-Blog world) through a posting at Joanne Jacob's site, where reader response has been strangely muted.  Arguments like this tend to bring out a "spirited" response.  I'd urge readers to go read his piece carefully -- it's well-written, thoughtful, and persuasive.   While my initial reaction to his argument was to reject it out of hand, the more I thought about it, the more convincing it became.  It's the sort of argument that deserves some serious thought.

Unfortunately, it's also a not a very good argument.  And I'd like to spend some time explaining in what respects I think it fails.

Finn starts from the highly questionable assertion that "Education is both a public and a private good."  It's a sweeping statement, but he quickly clarifies what he means:
Education is both a private and a public good, which is why states have assigned themselves constitutional responsibility for educating their young people. Every single state. * * * *  That’s why states have enacted “compulsory attendance” laws. And that’s why paying for education is the largest or second-largest item in the budget of every single state.
So Dr. Finn does not mean to make a blanket conceptual pronouncement about education, but rather to comment specifically on the sort of education provided by institutional schooling in our particular society.  That is, to the extent that Education is a "good" (and I take it he means this word in the sense of a commodity, and not in the sense of a virtuous telos) it is a good that has value and benefit both for the individual who receives this schooling, and derivatively for society at large.

Now, Dr. Finn's focus is firmly on the public aspect of schooling in this piece.  His treatment of education's being a private good is both peremptory and somewhat cryptic:
Yes, the private side of the education “good” is well represented, too. The Supreme Court famously ruled in 1925 that “The child is not the mere creature of the State; those who nurture him and direct his destiny have the right, coupled with the high duty, to recognize and prepare him for additional obligations.”
 This passage bizarrely seems to confuse the cui bono of education (which I took it was his primary concern and what characterizes education as a "good" with respect to one party or another) with the quo regor of education, which is purely a question of politics.  And further, it's not clear how the private side is "well-represented" or to whom, or why being well-represented means anything to the current discussion.  I think we can assume that what Dr. Finn is doing here is attempting to provide cover for the rest of his argument by acknowledging that there are interests in play beyond the public interests upon which his argument will depend.

Now to the meat of the argument, which shows up in this passage:
But when [parents] expect the state to educate their children at public expense, the public has a right to know whether those children are learning anything (no, not whether Johnny and Mary are learning, but whether the children of Waco—or Scarsdale—are learning); whether taxpayers are getting a decent ROI from the schools they’re paying for; and whether their community, their state, their society will be economically competitive and civically whole in the future as a result of an adequately educated populace.

The argument hinges on the idea that parents avail themselves of a benefit from society, so they must therefore subject themselves to all the ostensibly reasonable requirements that come along with that benefit.  If you want society to pay for your kid's education, well... society also has things it wants.  It wants to know that its money isn't being wasted, and it can only do that through testing.  So if you want education, you have to take the testing, too.  His argument seems to hinge not on the fact that parents have an ab initio obligation to testing, but that it arises as a voluntarily assumed obligation as a result of their participation in publicly-funded education.

This is a silly argument.

The notion that parents are choosing to avail themselves of public education is of dubious provenance.  Finn has already recognized, supra, the existence "compulsory attendance" laws.  Couple that with the existence of taxes collected to support the "largest or second largest" item in most state budgets, and you could easily argue (and I would argue) that it's rhetorically dishonest to talk about parents expecting to educate their children at public expense.  They are more or less forced to educate their children at public expense, having been taxed out of the ability to fund it themselves, and being further required to provide schooling at risk of having those children taken away.

Now there's a way to counter this argument.  Perhaps you don't see it like that: perhaps you understand and recognize that the parents are society, and that they have chosen to institute a series of laws that allow their children to be educated at public expense.  Now the parents are in fact making a choice.

Which would be all well and good for Finn's argument -- except that by transferring this responsibility to the parents, you also eliminate the idea that the parents somehow owe it "to society" to participate in testing.  The parents, as society, can choose to have whatever sort of educational system they want, and if the system that they produce is one in which opting out of testing is a legitimate choice, when that's what they "owe" society.

Of course, faced with these truths, Finn might simply respond that society has a right -- completely separate from the parent's willing participation -- to test children. He might simply jettison that part of his argument and say that there oughtta be a law.  And there is, sometimes.  Finn, to his credit, admits that the question of whether parents can legally opt out is a "murky" one, varying from jurisdiction to jurisdiction, and highly dependent on sometimes inconsistent enforcement.  He also recognizes that the question is not just legal, but political:
In the months ahead, states will have to clarify what is and isn’t required and how test participation is to be enforced—although, as with home schooling, they may choose to leave it murky rather than inviting big political battles
 And that's the real question: do we want to throw parents in jail and/or take their kids away for choosing not to have their children tested?  I doubt that the political will exists to really do this -- and I personally think it would be a terrible thing to do anyway.

But even if we did pass a law or series of laws that required testing, that still wouldn't mean that somehow the parents had voluntarily subjected themselves to the obligation of testing by "choosing" to have their kids educated at public expense.  That line of argument is simply vacuous.  Either the parents aren't really choosing, vis a vis society, or they are choosing as society and they have the right to choose whatever they want to choose.

But now we come to the real sticky thing about testing regimes, about which I've written before: even if you manage to muster the political will to enforce criminal penalties against anti-testing parents, you can't mandate that the children put forth their best efforts.  I mean, you can mandate it, but you can also command the waves to stay off the shore.  There is no practical way to prevent parents from instructing their kids to mark everything "C", or to fill in bubbles at random.  Lord knows I've deliberately tanked my share of tests in my life -- at least once or twice out of nothing but sheer spite.  And I didn't even have my Dad's encouragement.  I also know I'm not the only one. Testing requires active and willing participation to be of any use whatsoever.   To force academic testing on a population over protests is to critically affect the very thing that you're trying to measure.

In light of the fact that he's just wrong about the obligations of parents, and in light of the fact that forcing testing on people through political will is inherently unsound, I don't really understand Finn's heated rhetoric in the closing of his essay:
Our schools need to become more effective and our children need to learn more. Test results advance the public interest much as vaccinations do. (Maybe your kid is healthy today but the classroom needs everybody’s kid to be inoculated lest an epidemic start.)
Better tests are coming, but that doesn’t excuse “opting out” now. It’s not a legitimate form of civil disobedience. And it’s probably not legal, either. If you really find state tests odious, put your money and time where your mouth is—and stop asking taxpayers to educate your children.
Testing is nothing like vaccination.  At best, you can say that the institution of public schooling is like vaccination insofar as one educated imbecile really can ruin it for everyone -- and even that's a somewhat strained analogy. 

Testing either tells us something about the particular student being tested, the student body as a whole, or both.  (It may also tell us something about particular sub-groups in the student body, depending on statistical validity.)  Well, if what we're worried about is the individual, then I don't see the problem with opting out -- and Finn isn't making that argument anyway.  He explicitly stated that he wasn't interested in whether Johnny or Mary was learning, but that the public had an interest in knowing whether the children of Waco or Scarsdale were leraning.

But if what we're worried about is the statistical validity of the tests for the schools as a whole, well, then I'm almost equally flummoxed.  If 10% of students opt out of testing, you're still testing 90% of students.  You're still getting a good sense of what 90% of students are up to.  Do you know how many people Gallup talks to to get a decent model of the country's opinions?  A couple thousand.  Out of 300 million.

Rhee says (and Finn refrains) that testing is like stepping on the scale to find out if your diet is working.  Well it is.  But you also have to choose to step on the scale.  And you also don't need to step on the scale to see if the diet is working -- sometimes all you need to do is wake up and look in the mirror, or go for a walk and see how your knees feel.  Likewise, you can get a good intuitive sense of whether a school is doing its job by having a day's worth of conversations with its students.  (I'm not saying you get statistically sound quantitative data that way, but such data are not the only useful or legitimate epistemological objects.)

I want to be clear that I agree -- on purely pragmatic grounds -- with Finn (and Rhee) that testing is, like any sort of QC process, useful feature in our industrial education system.  Industrial-model schools can be supervised and sometimes made to work better when we can see what they are up to, and testing is the most efficient way to accomplish this.  (I do not agree that it is the best way, as Finn argues, but rather that it is the most efficient: that is, the best for the amount of resources and time that must be expended.  The best way would likely involve a prohibitive number of individualized observations and interviews.)  I'm also willing to accept arguendo that tests are getting better --  though I'm not convinced of that as an empirical matter.

But I've worked in a factory.  They don't measure every widget.  They measure maybe every 20th widget.  They break out the calipers, and clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-clink-MEASURE!

I suspect that what really bothers Finn is that the students who would be opting out of testing are, by and large, going to be drawn from among the most capable students.  In other words, by taking themselves out of the testing pool, they will make their schools look worse, on average, than they would otherwise.  (Assuming those same students had participated fully and actively in the testing.) 

But so what?  If the schools are only giving a good education on average, then they're failing in their mission anyway.

And if enough parents object to the testing that opting out does become a statistical problem, well, you're never going to get the requirement passed in the political arena anyway (assuming proper functioning of the democratic republican system, which may be asking a lot).

Which brings us back to Finn's primary, moral arguments: parents owe it to the schools because of the choices they made.

But that argument isn't a good one.

1 comment:

Reine Marie said...

This is not an either/or question. The benefits of college are not all public and they are not all private. The debate should be over what fraction of the cost of postsecondary education students should bear and how large society’s subsidy to them should be. It should not be over whether education is a “public” or a “private” good.