For the last week, we have been following the debate in conservative educational circles set off by Diane Ravitch’s counterfactual on teacher unions and student achievement. If teacher unions have the negative effect on student achievement that anti-unionists never tire of postulating, Ravitch asked, why is that that the state with the highest academic performance, Massachusetts, is also one of the states with the greatest teacher union density and strength?
Ravitch’s counterfactual was so bedeviling in these circles because it employed an Ockham’s razor on the thicket of prejudices about teacher unions that underlies much of conservative educational discourse on the subject. If we accept the hoary first principle that teacher unions have a clear and unambiguously negative effect on student achievement, her argument went, it should follow logically that the stronger teacher unions are, the poorer student achievement will be. The fact that the converse holds true should cause some pause and critical scrutiny on the validity of that first principle.
Ravitch’s counterfactual works well as critique, but it does not offer an alternative theory of the relationship between teacher unionism and student achievement. From a social science viewpoint, there is very good reason to be deeply skeptical about our ability to establish a meaningful causal relationship between teacher unionism and student achievement, as we pointed out in this previous post. Still, it is still worth our while to analyze the arguments put forward on behalf of the claim that teachers unions have a negative effect on student achievement, because they provide some insight into the impoverished thinking that underlies this point of view.
Among those in the edu-blogophere who see teacher unions as always and everywhere a nefarious influence, the response to Ravitch’s counterfactual can be distilled into the following two conclusions, simultaneously embraced:
- Teacher unions are so strong that even in ‘right to work’ states like Mississippi which allow no collective bargaining they are able to impose their will and their work rules, leading to poor student achievement.
- Teacher unions are so weak that even in states where their density is well over 90% and they have their best organized political presence like Massachusetts they play no role whatsoever in good student achievement, which is accomplished in spite of them.
So teacher unions are powerful behemoths where they are the weakest, and paper tigers where they are the strongest. That is the sort of convoluted reasoning which was explained away in the past as dialectics by those whose knowledge of philosophy began and ended with Mao’s On Contradiction.
Viewed in this form, the nature of the discourse being employed becomes plain: what we have here is actually a form of stereotyping. Stereotyping thought is well-known for its propensity to hold directly contradictory claims about its subject, with the only unifying element being that the competing precepts all put that subject in the most negative light. [Interestingly, Sherman Dorn seems to have reached a similar conclusion, in a post published as I was finishing these thoughts: he describes the discourse as "demonization" and "scapegoating."] In anti-Semitic stereotyping, Jews have been commonly held to be both greedy, exploitative capitalists AND ruthless communists out to seize all private property. Similarly, in homophobic stereotyping discourse, gay men are portrayed as both weak and ineffectual AND powerful sexual predators who threaten the most basic social institutions. The examples of such contradictory stereotypes are legion, and are highlighted in social science analysis such as Gordon Allport’s classic on The Nature of Prejudice. In classic stereotyping form, Fordham’s Mike Petrilli “reconciles” the behemoth and paper tiger contradiction with a summary judgment on teacher unions as the enemy of all that is educationally good and right: “they are tenacious and need to be defeated, over and over and over again.”
If one wonders why educational conservatism seems so bereft of interesting and compelling ideas these days, the nature of the discourse employed in this debate would be a good place to start.
5 Comments:
1 Peter Hilts
· May 30, 2009 at 3:18 pm
This is almost comical. The plain and undeniable truth is that educational reform in MA happened despite strong and persistent resistance from the unions. Ravitch conveniently ignores this fact and gives the unions causative credit where not even a correlation exists. Ockam’s Razor doesn’t work when the argument ignores basic facts.
Observing that unions both bluster and whine is far from a contradictory stereotype—but when the facts don’t support your argument, its always easier to mischaracterize the opponents’ claims.
2 Leo Casey
· Jun 1, 2009 at 2:56 pm
Little surprise that those on the ultra-right of the educational blogosphere, such as Peter Hilts, feel a need to misrepresent what Diane Ravitch actually wrote. I would simply encourage readers to follow the links in the original posts, and read for yourself.
As for the “plain and undeniable truth,” I would simply note that no one on the teacher union side denies that we oppose the sort of educational reform put forward by Hilts and his friends — we do oppose efforts at privatizing public education and we do oppose the excessive use and abuse of standardized exams. What we hold is that there is no evidence demonstrating that either policy improves education, in Massachusetts or anywhere else.
3 Peter Hilts
· Jun 4, 2009 at 9:09 pm
Thank you Leo. You disagree with my comment, so you immediately reveal the weakness of your argument by going ad hominen. Nicely done.
Actually Leo, I am correctly representing what you wrote. Ravitch repeatedly implies a correlation between high unionization and high performance. Please refer us to any mention of union opposition in the original post. Ravitch is far too smart to ignore her history, so it is a reasonable inference that she excluded relevant information on purpose.
Now—in defense—she has backed of her correlation argument and instead asserts that unions don’t block reform. That’s quite a retreat.
Since MA was highly unionized before reform, and remains highly unionized after reform, what accounts for the improvement? It is those very reforms that you oppose.
In the presence of a change in the dependent variable, if one independent variable remains the same, while another independent variable changes, the simplest explanation is that the changing variable is responsible for the change in outcome. That’s a correct application of Ockam’s razor.
4 Leo Casey
· Jun 10, 2009 at 7:37 pm
Peter:
1. You have a rather liberal notion of ad hominem argument.
2. You also don’t seem to understand the distinction between correlation and causation. Yes, Ravitch says there is a correlation between high academic performance and strong teacher unions in Massachusetts. Both exist at the same time and in the same place, which is what qualifies as correlation.
What she doesn’t say is that there is a causal relationship, that strong teacher unions cause high academic performance. That is a different claim, with an entirely different status. There are many correlations that clearly do not qualify as causal.
Lots of things are correlated with high academic achievement. Take geography, for one example. On the whole, northern states have a higher academic performance than southern states. But no one would seriously argue that the geography is causal. Rather, there are historical reasons why northern states are better performers.
3. Massachusetts was always a high performing state. What changed in the last decade was that it went to the very top of the pack. You leave out of your account a number of very plausible reasons why that might have happened. For example, Massachusetts had a successful adequacy lawsuit, supported by teacher unions and opposed by the sort of privatizing reformers you so admire, which dramatically increased the revenues and resources going to public schools, and especially those schools serving the neediest students. As an explanation for moving from the top 5% of states to the very top, that is a whole lot more plausible than your claim for increasing the numbers of standardized tests.
Leo
5 Peter Hilts
· Jun 15, 2009 at 5:02 pm
Leo, I thought this comment thread was dead, but I’ll chime back in. You are welcome to have the last word. I’m hopeful we’ll continue to dialogue and debate in other settings. I’ll cross-post my reply on my blog so more folks can extend the discussion.
1. It is ad hominem (and silly) to characterize me among “those on the ultra-right.” You are attempting to impeach my credibility by using a shortcut label. That isn’t a response to my argument, but to me as a person—that’s exactly what the conventional (not liberal) interpretation of “against the man” means. Whether you did it thoughtfully or reflexively, your introductory salvo remains ad hominem. Your writing and public testimony is full of loaded language like “privatize” “dismantling” etc. When those you support have a plan, you call it a “proposal”, when you oppose a plan you call it a “scheme”. Your use of language is nuanced and persuasive, but don’t pretend it doesn’t reveal a heavy bias. In the circles where you live, work, and publish, labeling someone “ultra-right” demonizes them.
2. Thanks for the primer on inferential statistics. Ravitch does cross the line from correlation to causation, and she does so explicitly. Why else would she reference both Finland’s rate of unionization and their high performance on PISA? In fact, Ravitch explicitly offers her analysis in response to the argument “teachers’ unions were no help to education reform.” That phrase “no help” is a causative argument, not a correlation. It may be subtle, and it may be poorly warranted, but Ravitch is attempting to refute a causative argument (of her own straw-man phrasing) with a correlational argument. Are you suggesting that Ravitch thinks unions are neutral? She is both making a rebuttal argument and claiming that no argument can be made.
3. Here you make a legitimate argument about which changes are driving improved performance in MA. I don’t know that answer, but like many dynamics in education, there are probably multiple causes. That will take much more rigorous research than currently exists. (It is disingenuous to pretend that I have advocated some increase in the number of standardized tests, or that the reforms in MA are just about testing—please substantiate or drop that line of [reasoning]) What is obvious though, is that improvement in MA has occurred in an environment of reform, and that the reforms in MA happened over the robust objections of the unions. I am convinced that if the type of charter school you advocate elsewhere had been proposed in MA 15 years ago, the union would have opposed it, calling it a “scheme” of the “ultra-right” to “cherry-pick” students, “drain resources” “privatize” public education etc.
Ultimately, the existence of charter schools will make public education better. The presence of healthy and balanced assessment systems will make public education better. The healthy checks and balances of teacher unions will make charter schools better. Teacher resistance and potential for decertification will make unions better. What won’t improve the situation is muddled argumentation and attacks on messenger with whom you disagree.