Bottom 5%

Five percent.

It's a figure that turns up again and again in reformster rhetoric, usually teamed up with the word "bottom."

It has a fine long history. All the way back in June 2009, we can find Arnie Duncan talking about the five percent in his address to the conference of the National Alliance of Public Charter Schools. The address, "Turning Around the Bottom Five Percent," and it features the rhetorical sleight-of-hand that usually accompanies discussion of the five percent. Duncan leads with a description of chronically under-performing schools, noting the social and physical conditions of these schools are "horrific." "They're often unsafe, underfunded, poorly run, crumbling, and challenged in so many ways that the situation can feel hopeless."

That was six years ago. Since then, the five percent have been cropping up regularly. Michigan is just one example of a state that targeted the bottom five percent of schools for turnaround and takeover. The speading-like-a-slow-ugly-weed program of Achievement School Districts (ASD, or "sad" spelled sideways) love the five percent figure. In Tennessee, Chris Barbic promised that he would take the bottom five percent of schools and lift them into the top twenty-five percent (spoiler alert: Barbic didn't even come close, and is instead getting out of Dodge). Meanwhile, in Pennsylvania, Georgia and North Carolina, legislators are talking about adopting the ASD model, complete with "bottom five percent" qualification for takeover.

The "bottom five percent" meme is firmly ensconced in the language of reform. Just today, we find Hunt Lambert, the Dean of Continuing Education and Extension, opening a piece on his life-long learning blog with this sentence:

Improving the bottom five percent of K-12 schools is one of the most vexing problems in America. 

Setting aside his use of "vexing" (is vexation the next step of agitation past pearl clutching?), here's one more presumably educated person talking about the bottom five percent as if it's an actual thing.

There isn't a shred of evidence to suggest that the bottom five percent of schools have some particular characteristic that makes their turnaround, takeover, or general state nose-poking are required or effective. It could have been the bottom six percent, or four percent, or eleven-point-oh-two-six percent. Five is a just a nice, round, simple number. The fact the everyone from California to Oklahoma to New York is hunting down the same five percent figure could be a sign that five percent is a very attractive number, an appealing number, the Younger Brad Pitt or numbers. Or it could be (though I haven't found direct evidence of this, yet) that all these states are working from the same template, perhaps one created by ALEC?

So is the number bad just because it's arbitrary?  It's a fair question, and the fair answer is, no, no it's not. After all, we love arbitrary-ish performance measures in education. Should the cut-off for an A be 90% or 93%? If you've worked in a building that had the grading scale argument, you know just how fact-and-data free that discussion can be.

But no-- there are other, larger reasons that the "bottom five percent" cut-off line is wrong, and bad, and should be avoided like the plague.

First, that "bottom" is troubling. You remember when Arne painted a picture of chronically under-performing schools with "horrific" conditions. But "bottom five percent" means just that. If we go to a wealthy county where the schools have great resources and support and the graduation and college completion rates are sky-high, somewhere in that district there are still schools that fall in the bottom five percent. If we look at the prestigious pre-ivy private prep school network, somewhere amidst Trinity and Philips Exeter, we'll find the bottom five percent. If the students in my class all master a skill and ace the test so that the grades range between 95% and 100%, someone in that class is still in the bottom five percent.

If we set a bottom standard and said, "All schools must come in higher than this cut line," there could conceivably come a day on which we said, "Mission accomplished! Every school has made the cut. We win! Let's declare victory and come home." But there will be a bottom five percent forever, and so the reformster's work will never be done. There will always be schools targeted for closure and takeover. Always.

Second, this is stack ranking. Stack ranking enjoyed a vogue in business and industry, but the curtain fell loudly upon the practice ranking employees in comparison to each other and firing the bottom of the stack. Most famously, Microsoft dropped the practice in 2013, not because it was philosophically repulsive or made people sad, but because it just didn't work. The New York Times Business section just last week attacked the practice once again reporting on yet another study showing that stack ranking doesn't work and in fact does more harm than good.

And those problems, both serious indictments of the "bottom five percent" approach to reform-- those problems are before we even get to the question of how schools are ranked in the first place. The answer, generally, is "by test scores." So those same old Big Standardized Test scores involving badly designed questions aiming to measure a very narrow sliver of student ability will somehow tell us which schools are better or best.

Or the other big question mark in this whole system-- the state will take those bottom five percent schools and do.... what? Turn them around and fix them? Is there any indication that the states or the privateers that they invariably hire to do the work-- do any of them know the secret sauce for turning schools around? If they don't, then what is the point of this exercise? If they do, why did we decide that only the bottom five percent would get the benefit of this miraculous brew of fairy dust and unicorn pee?

Any time you see "bottom five percent" crop up, beware. It's one more time that reformsters are just making stuff up but trusting you'll believe them because, look, numbers!

UPDATE: Somehow I neglected to include the "research" from Chetty, Friedman and Rockoff that canning the bottom 5% of teachers (identified using magical and debunked VAM sauce) would make schools more awesome. Such a magical number, that 5%.

The GOP vs. DEM Question

Discussion surrounding ESEA re-authorization has helped clarify the question that distinguished between the Democratic and Republican parties when it comes to education--

Is it better to have stupid education policy implemented by the state or federal government?

Democrats believe (and asserted it with vigor in their ill-conceived Murphy Amendment) that high-stakes testing and prescriptive punishments attached to test results should be managed by the federal government. The GOP, however, argues that the feds should be left out of the equation and that power brokers on the state level should have the privilege of crushing public education and selling off the parts to privateers.

Neither party has displayed any interest in discussing what educational policies might actual help public education in the US. Neither party has shown an ability to hear the chorus of educational experts and qualified researchers pointing out that stack ranking, magic VAM sauce, and disenfranchising local members of the community are not working and will most likely never work.

Neither party has displayed an inclination to say, "Hey, we've been doing some of these things for fifteen years now. Are there any signs that any of it actually works?"

Nope. The parties disagree one one point only-- should the feds be in charge, or should each individual state run its own show.

That's our choice, apparently.

Given that choice, I lean toward the state-driven answer.

Not that I have unbridled faith in the states. Guys like Scott Walker and Andrew Cuomo have shown themselves completely capable of destroying their state's public education systems without any help from the federal government.

But I side with states for two reasons.

First. the state can only hurt the state. Cuomo can make a mess out of New York, but his idiotic ideas have not done any damage here in Pennsylvania. Sam Brownback may be able to shamblize education in Kansas, but Arne Duncan has been able to screw up education in all fifty states.

Second, on the state level, progress can be made. In Pennsylvania, we managed to can a governor who was destroying public education and replace him with a governor who is at least trying to help. Local activists can focus resources on local issues, while the national-scale corporations and activists who are pushing the reformster agenda have to fight on fifty fronts.

It isn't optimal. Optimal is elected officials and policymakers who listen to educators and use their powers of critical thinking to figure out the obvious-- that the policies which reformsters love are bad for education, corrosive for democracy, and don't actually accomplish any of their stated goals. That's the conversation we need to have. But in the grand debate between DEM and GOP, only one question is on the table--

What's the best way to implement bad policy? Is it better to have stupid education policy implemented by the state or federal government.

Test & Punish & Civil Rights

The Murphy Amendment (Premise: the worst parts of NCLB are actually the best parts; let's give them steroids) was one more manifestation of the Civil Rights Argument for Test and Punish. Amendment proposer Chris Murphy (D-Conn) invoked that justification for his test and punish bill, saying that ESEA "has to be a civil rights law."

Kati Haycock is president and founder of the Education Trust, a Gates-funded test-pushing advocacy group that supported NCLB and helped craft the Common Core. She took to her website to call Lily Eskelsen Garcia a liar for claiming that test and punish policies are not a civil rights win. Haycock is just one of the many civil rights advocates who speak out in favor of test and punish. But there is now a large coalition of civil rights advocates who speak out against test and punish, as well.

I've written about this again and again and again and again. But as the reformsters have found a strong tactical advantage in using the civil rights argument to promote test and punish, let me see if I can distill the important points of the argument here.

There Is a Real Concern

The systemic ignoring, underserving, and general neglect of non-white, non-wealthy populations is a real problem. "Do nothing" and "Go back to doing what we used to do" are not viable solutions.


Testing doesn't tell us anything we don't already know

Actually, test scores don't tell us much of anything, because the Big Standardized Tests are narrowly focused, poorly designed, and extremely limited in their scope. Furthermore, we can predict test score results pretty well just using demographic information. So to claim that we would be fumbling in the dark without these tests, with no idea of how to find schools that were in trouble, is simply ridiculous.


Nobody is sending help

Advocates argue that test scores provide political leverage that forces The System to respond. But from New Jersey to Philly to Detroit to Chicago, the response has been the same-- instead of help, politicians silence and disenfranchise the members of the community and privatizers come in to strip-mine the community for profit.

If policymakers responded to low test scores by sitting down with community members to say, "How can we help you," and channeling resources to the schools, I'd feel differently about all this. But that's not even sort of what's happening. Instead, charters may "save" a handful of students while simultaneously making the public schools even worse for everyone else.

P.S. Watch for cynical bullshit 

The Murphy Amendment included the famous "bottom 5%" clause, which is always a tip-off that vampires are at work. This is a guarantee that there will always be fresh meat, because even if all the schools in your city score between 95% and 100%, there will still be schools that are in the bottom 5% of that.


Test and Punish is a False Narrative

The claim is that there is a plane crash out there, and we can only find the victims by releasing the hounds. But the hounds are not rescue animals, but blind and deaf hunting dogs, and the smoke from the crash is clearly visible in the sky, and instead of sending doctors and rescue personnel, the powers that be are sending vultures while beating the doctors for not trying hard enough.


It's doubly anger-making because there are people who really do need our help, but instead of trying to actually help them, we've got people gaming the system in order to profit from the problem. This is not okay. And that's as simple as I can make it.

Public Ed Needs Allies

With the now-thankfully-defeated Murphy Amendment, Senate Democrats gave a giant middle finger to public education and a bathtub full of cold water in the face to those who keep thinking that maybe the Democrats in general and Progressives in particular are going to be our allies in our struggle to preserve the promise of public education.

They aren't.

Steven Singer lays out the shock and dismay pretty clearly.

Up until now I’ve always been with the Democrats because they had better – though still bad – education policies than the Republicans. I’m not sure I can say that anymore. In fact, it may be just the opposite.

So what's the fuss? The Murphy Amendment was an attempt to put the test-and-punish back into ESEA, including solidifying that magic "bottom 5%" rule into federal law. It was a way for Democrats to say that they actually loved them the last fifteen years of test-and-punish based ed reform and they would like still more of it. And it took the GOP to stop these dopes.

This is not entirely a shock. The Democrats have given plenty of notice that they are not friends of public education, not the least of which would be two entire Democratic administrations under Obama-Duncan. I know die-hard Dems like to imagine that Obama is some sort of outlier or that Duncan is a rogue Education Secretary, but the sad truth is that a Democrat has had the chance to set education policy, and that's what we've been living with for seven years. The Murphy Amendment doesn't represent a new shift or alliance or change in direction. It's right where they've been headed all along.

The only bright spot in any of this was that the NEA was vocal and on the right side of this and not making nice with the Democrats (because, hey, they're our political allies).

I do not know the answer to the political calculus of public education in this country, but I do know that we have got to stop blindly supporting parties and start focusing on policy. And we have got to stop pretending that the Democrats are our friends no matter what. For that matter, we need to start distinguishing between good relationships and good policies. The fact that we may have a "good relationship" with Democrats does not mean they won't screw us, and the fact that we have a "bad relationship" with some Republicans does not mean that they won't support policies that help public education work better.

Public education is a political orphan, with few politicians watching out for us. The Murphy Amendment is just one more reminder that just because you think someone's swell, that doesn't mean they stand for what you wish they stood for.

It is a dark part of Democrat dna to think that only filing reports with the feds makes the world go around, just as it is in the dark part of GOP dna to think that those who can't pay the price of admission to society's lifeboat should just be left to swim home on their own.

Both the GOP and Dems are infected with money. Neither can be trusted as a group. Certainly neither can be trusted all the time to watch out for American public education, because neither party believes in the promise of public education any more. I'm not sure what the answer is. Take personal responsibility for getting the word out (don't just say "Well, I pay union dues so they'll take care of it). Contact your representatives early and often. Convince others to do the same. Raise a fuss and make some noise. Pay attention ALL THE TIME. And remember that this is a marathon, not a sprint.

Public education needs allies. I suggest that rather than farm the job out, we start with ourselves, and make ourselves into the allies that public education needs, because the folks in DC sure as hell aren't going to do it.

Can Third Graders "Fail"

A twitter conversation this morning turned into a discussion of the semantics of talking about test results, but any conversation that turns to phrases like "semantics of talking about test results" (39 characters) is kind of doomed on twitter. Jennifer Borgioli (@JennBinis) referred me to her article from last summer addressing the issue, "The Semantics of Test Scores."

There's a fairly detailed illustration of her argument, but her general point is that people play pretty fast and loose with the term "fail," particularly with the 3-8 grade range of tests.

Generally speaking, when we talk about a test where the resulting score is described as passing or failing, it’s in relation to the consequences for the test taker. Fail your driver’s test? You can’t drive. Pass your boards? Welcome to the profession, Doctor. 

While that holds true for many high school Big Standardized Tests that are used as graduation requirements, Bergioli argues that no such "bright line" exists for BS Tests for grades 3 through 8. 

There are no short-term negative consequences for students in grades 3-8 based on their performance on the state assessments. 

Bergioli's example is rooted in New York; this "bright line" assertion is, of course, flat out false if we throw in states like Mississippi that like the idea of holding back third graders who score too low on the reading assessment. That, I think, qualifies pretty clearly as failing.

I do get her point. Language choice with children (particularly younger ones) is important, and it is particularly important to choose carefully when discussing success or the lack thereof. When my kids were little, their mother and I were careful to use phrases like "haven't succeeded yet" in place of "failed." When we designed graduation projects for my high school, the only outcomes we made a place for in the evaluation stage were "successfully completed" and "not successfully completed yet." 

I get that some test-loving reformsters imagine a perfect world where tests are given, tests come back, and nine year olds say, "Well, that was not what I had hoped for. But I can see that I need to enrich my study and practice of identifying main ideas in paragraphs, so I guess I'll just hunker down and do that." 

But in this, as in so many areas, folks who design and promote this stuff are kidding themselves.

So, while there is once again the possibility of larger, longer-term consequences to the school, the district, and the community based on how a group of students do on these tests, in the absence of a clear and bright line of a relationship between student performance on the test and consequences to the student, it’s misleading to say the student “failed” the test.

No, I don't think it's misleading at all.

First of all, the larger, long-term consequences are based on drawing a clear and bright line. The state says, "This number of your students fell on the wrong side of this clear, bright line, so you are not an effective teacher." Depending on your state and its laws, that ineffectiveness may be reflected in your evaluation, your pay, and your future employment. It will also be reflected in the rating of your school, and of course in many states that rating will be a nice, neat letter grade. Why do some policymakers like giving schools letter grades? Because it makes it easy to tell if the school is passing or failing.

Second, children are not dopes. Decades of sorting students into bluebirds and chickenhawks have fooled almost nobody-- students know whether they're winning or losing in the Big Game O' Learning Stuff. And because the stakes on test results are so high for schools, students who score Not So High on BS Tests will find themselves rewarded with extra work, extra practice, extra time in the albatross reading group. It can take a tough little kid to look at the evidence and not reach unflattering conclusions about herself, and most teachers I know do their best to keep children from reaching those conclusions. But students know when they've failed, whatever we try to call it.

We can say that these students have not technically failed, and in an academic technical sense, we are correct. But eight year olds are not known for their ability to look at things in an academic technical sense. That's one of the truly toxic effects of badly written tests-- young students lack the capacity to say, "Well, this was a poorly designed assessment." They just think, "I must be stupid."

Nor do we get a lot of nuance from the policymakers and politicians who keep talking about failing schools and failing teachers, all of which underlines clearly that there's a bright clear line, and anybody falling below it has failed. 

Test manufacturing experts sometimes remind me of sad scientists in old SF movies. They design these instruments to try to tease out nuanced granular pictures of student strengths and weaknesses and then policymakers just grab the tests and say, "Never mind all that. I just wanna know how many of these kids failed."
 

I agree that it would be better for everyone if we could deal with these issues in a nuanced thoughtful manner. But then, if the education discussion were being run by policymakers who valued nuance, detail, and the expertise of people in the field, we'd be in a far different place than we are today. We can try to shade the meaning of "fail" so that it has a very specific meaning in very specific circumstances, but that's a hopeless exercise. Everyone knows what "fails," means-- you came in below a particular mark. And that definition fits for every test in the history of ever.

 

How AFT Blew It

I wouldn't devote one more post to deconstructing Randi Weingarten's early-bird Christmas gift to Hillary Clinton except that I'm an NEA member, and I'm living in fear that NEA president Lily Ekelsen Garcia's administration will lead us down the same path.

Arguing about Clinton (or Weingarten for that matter) is a tricky dance. Clinton tends to have a destabilizing effect on the brains of people who don't like her, who proceed to froth at the mouth at start ranting about conspiracies tortured enough make a truther blush. My opposition to Clinton (and support for Bernie Sanders) is not based on any belief that she is a terrible human being, a crazy-awful person, or some evil mastermind bitch on wheels. My reluctance to support her is not even based on my perception that she is extraordinarily inauthentic (though I think that magnifies her other issues). I just don't think she is remotely a supporter of public education or the teachers who work there. I think she would be perfectly comfortable continuing the exact same policies that we've suffered under for the past fifteen years and in fact would prefer to continue with them.

The counter-argument is that she's electable while Sanders is a modern George McGovern, beloved by liberals and doomed in the general election. Maybe Clinton is electable (though if that's the case I'd ask why? Could it be that she's electable because, other than her lack of a penis, she is indistinguishable from a Republican candidate). And there's a case to be made that endorsing early and ahead of the pack earns you a better voice in policy discussions.

But if she is electable, nobody's pretending it won't be a tough sell. And that's how AFT blew it.

Let me take a moment to tell you how I have always handled advising students who are in charge of putting on Prom. First, it takes months-- mooonnnnnnths. Because we make sure every student on class council has had a chance to propose an idea and explain that idea. Then students break into groups and they research and pitch the ideas. And then they discuss the ideas until every single person has been heard just as much as they want to be. And then they decide.

The process is long and involved and inefficient and often results in exactly the same theme-and-decoration decisions that the class president or I could have simply installed by fiat at the beginning of the process. But the long involved process doesn't just result in a decision about which color of vinyl to hand behind the cardboard castle. It also results in an entire group of students who are energized, informed, and invested. They know what we're doing, why we're doing it, and why it's the best solution we could come up with. They will work for that theme, even if it wasn't the one they most wanted.

The process of deciding on a candidate to endorse could-- and should-- also be the process of getting members educated, involved and invested in the decision. But leaders of large, unruly, cat-herding groups like a teachers union are reluctant to relinquish control. The irony here is that such groups are often run with the same top-down management style that has helped make education "reform" such a train wreck.

While AFT's tiny sample was legit for sampling purposes, Daniel Katz correctly notes that the massaging of the data makes it a bit suspect. But what it mostly did was completely fail to engage the members, and now instead of delivering a groundswell of Clinton enthusiasm, Weingarten delivers a free-wheeling cat-herding argument about leadership's choice for the union. Instead, the people who are excited about the endorsement are people like DFER-- the faux Democrats (also, faux democrats) who would like teachers to be Put in Their Place and for the unions to die. That's who's excited about this.

I believe some folks have grossly over-estimated Clinton's electability, under-estimated Sander's electability, and hugely under-estimated how much Clinton really doesn't support public education and the people who work there. I suppose time will tell.

But in the meantime, I'm really, really hoping that NEA will take a more careful approach to an endorsement. I hope we don't send the Dems the message that we will always be there for them, no matter how badly they treat us. I hope we don't cut the membership out of the process and just expect them to fall in line. And I hope we endorse somebody who isn't going to, once again, stab us in the back, front, and side.

The AFT used a long questionnaire  (about twenty-six questions, only eight of which directly addressed education). I'd like to see the NEA's list of questions include these two:

What did the Obama/Duncan administration get wrong about education?

What would your administration do differently going forward?

Because any candidate that wants support from teachers ought to be able to answer both of those, clearly and specifically. I'll be waiting.

Sad Days in PARCCland

PARCC, built on the dream of a national scale standardized testing system, has been dumped by yet another state. Governor John Kasich of Ohio two weeks ago signed a budget that severs Ohio's connection to the PARCC consortium. The dream is dying.


The PARCC had only been used in Ohio for one year-- but that year was disastrous. State Senator Peggy Lehner (who may be a saint or who may be an opportunist who saw a political opening) had set up her own committee to look into testing issues, and what she found in a survey a few months ago was that basically every sentient human in Ohio hates the PARCC test. In her survey, Lehner found exactly one superintendent who "strongly agreed" that the implementation of the tests went well.

The Ohio legislature has been after PARCC for several months, including an earlier proposal that not only cut the test but cut the education department's budget for all testing (the new bill allows for new tests, but shorter and given only once at year's end). And way back in February, 25-year veteran teacher (and BAT) Dawn Neely-Randall spoke out against the test, helping kick off an avalanche of criticism from teachers.

Governor Kasich, a Common Core True Believer who has labeled Core opposition "hysteria," has tried to defend the PARCC, but clearly has given up that fight. His Presidential aspirations may or may not be a factor, but Kasich is the only other Republican besides Jeb Bush who would conceivably not run away from Common Core. For PARCC, this is more bad news.

When the testing consortium was launched, it included twenty-three states and the District of Columbia. Ohio's defection brings the number down to ten states, plus DC. Those states are Arkansas, Colorado, District of Columbia, Illinois, Louisiana, Maryland, Massachusetts, Mississippi, New Jersey, New Mexico, and Rhode Island (that's according to the PARCC website, which hasn't been changed to reflect the new loss).

You'll notice two things about the list-- one is that some of those states are not necessarily solidly in the PARCC camp, and the other is that they are not among America's most populous. PARCC was only testing five million students when Ohio was still in; now that number grows smaller still. It's no wonder that testing advocates are pushing so hard to keep every year, every student testing mandates in the ESEA rewrite-- the market and attendant revenue stream for test manufacturers like Pearson is plummeting. (Update: Let me clarify that-- the market for state-by-state manufacturing is growing, but the chance to blanket the entire country with a single test product is falling apart.)
In fact, the indispensable Mercedes Schneider reports this morning that marketing research firm Questor now lists Pearson as a "sell," citing in particular the growing scrutiny of testing in North America.

There's no longer any question that PARCC's golden days are long behind it. It will be interesting just how small the test can get before Pearson decides that the much-unloved not-so-mega-test is no longer worth their corporate time, trouble, and investment.

This piece ran just two weeks ago at View from the Cheap Seats. I'm happy to report that in the interim, things have not looked up for PARCC.