Friday, December 19, 2014

Cyber-Security and Our Data Overlords

In the wake of the hack of Sony Pictures over a Seth Rogen movie (because that's the world we live in now-- one where a middling stoner comedy prompts international incidents), comes this promo line from CNN:

Sony hack was bad, but worse is coming.

The story quotes the deputy director at the Center for Strategic and International Studies, Denise Zheng, commenting that cyber-espionage is happening at an unprecedented rate. Tony Cole, vice president of security firm FireEye said, "This is a global problem. We don't have a malware problem. We have an adversary problem."

The government's own report indicates 61,000 cyber attacks and security breaches. The director of information security issues at the Government Accounting Office is quoted as saying that federal tech is "not as secure as it should be. GAO has been identifying this area as high risk since 1997. It's been a longstanding challenge for the federal government to adequately protect its systems."

So the feds can't keep things safe. A multi-national corporation like Sony, with massive resources and a high motivation to keep their bread and butter in airtight lockbox, can't keep things safe.

Let's set this news next to Pearson and the rest of our Data Overlords whose dream is a vast data ocean so filled with information that it will be able to tell you what breakfast you should eat on the day you're taking a math test. The Data Overlord wing of Reformsterism envisions a day when the womb-to-workplace pipeline is so jammed with information-- both about a person's intellectual achievements and character traits-- that a quick look will tell us everything about that person (including what sort of cog-like function he can best fill in the world).

The intentions surrounding this Big Brothery technoventure are scary enough, but if we add the potential for hackage, pillage and general theft of the information, it becomes even more disastrous. Bad enough that the government and corporations want to gather a huge file on you in order to select your destiny for you. What happens when someone decides it might be profitable to steal your life and all that's in it.

The Data Overlords have periodically reminded us that they'll absolutely keep all that data secure. The news from Sony and the reactions to it remind us that there isn't a government or a corporation on the planet that is actually capable of keeping that promise. It's as if the Data Overlords are saying, "Look, just give us every valuable thing you own, your money, your jewelry, your ATM cards and the pin numbers with them, and we will put them all in a bucket, which we will sit out on our front porch. It will be totally safe. We promise."

Thursday, December 18, 2014

Cuomo Doesn't Understand Accountability

There are many levels on which Andrew Cuomo doesn't grasp accountability, including, say, the personal accountability involved in giving one's word and then not keeping it.

But his recent remarks about the evaluation and accountability system for teachers in New York State shows that he, like many other reformsters, just doesn't get the point.

Cuomo has been criticizing the results of the evaluation system because he believes they are wrong. This is backwards.

The whole point of any sort of assessment, diagnostic, or accountability system is to find out exactly what's going on. A doctor gives a cancer test to find out if the patient has cancer. The doctor does not get the test back from the lab and say, "No, this is wrong. I already know whether the patient has cancer or not. Keep doing this test till it confirms the answer I have already decided on."

The point of any sensible accountability system for teachers should be to find out how well they are teaching. There are many ways to assess whether a system is actually telling you the truth or not, ways to see if the answer you're getting is accurate. But Cuomo apparently believes he already knows the answer, and he is just looking for an accountability system to confirm what he already knows.

We have of course seen systems like this before, but they are usually employed to hide failure. In Communist China, the leaders decided that shifting to an industrial economy would not harm their food supply, and so reports from the farming districts had to be rewritten until they showed that farms were still producing more than enough food. Meanwhile, millions of Chinese starved to death. In Vietnam, generals demanded that field reports be rewritten until they showed that our troops were winning the battles. That worked out well.

Cuomo has turned history on its head. Instead of hiding failure, he would like the New York teacher evaluation system to hide success, and he will keep rewriting the system until it produces the reports of failure that he demands (whether they are real or not).

Governor Cuomo (and many other reformsters) does not understand the purpose of an evaluation system. A sensible human does not go to the doctor and say, "I know I have cancer. Confirm it or you're fired." A sensible human does not pull out a thermometer, "I already have guessed what the temperature is. If this thermometer doesn't get that answer, it must be broken."

The purpose of an accountability system is to figure out what's going on, not to play Gotcha. By insisting on an accountability system that is based on his preferences instead of actual reality, Cuomo positions New York as a state where teachers can be evaluated as failing for No Damn Reason. He will certainly not be alone in that, but if he wants to convince teachers to come work in the Empire State, he needs a better plan. I would recommend one based on reality.

Race to the Top Priorities

The giant turkey that is Race to the Top has its neck on the chopping block. I would not celebrate just yet-- a proposed budget is about as solid and secure as the sticker price on a used car. But at the very least, the CRomnibus bill is a shot fired across RttT's bow.

So that (and, you know, Throwback Thursday) make it a great time to go back to this document, the executive summary of RttT. The summary includes a list of judging criteria for RttT applications, including the point value for each one, and while they are arranged by categories, I thought it might be useful to arrange them in point order, form the most points to the least. As with any rubric, the point assignment reveals what the real priorities are. So let's see where RttT's heart really lay.

Articulating State's education reform agenda and LEA's participation in it (65 points)
Improving teacher and principal effectiveness based on performance (58 points)
Developing and adopting common standards (40 points)
Blah blah blah promoting charters (40 points)
Turning around the lowest-achieving schools (40 points)
Building strong statewide capacity to implement, scale up, and sustain plan (30 points)
Demonstrating significant process in raising achievement and closing gaps (30 points)
Ensuring equitable distribution of effective teachers and principals (25 points)
Fully implementing statewide longitudinal system (24 points)
Providing high-quality pathways for aspiring teachers and principals (21 points)
Supporting the transition to enhanced standards and high quality assessments (20 points)
Providing effective support to teachers and principals (20 points)
Using data to improve instruction (18 points)
Improving the effectiveness of teacher and principal preparation programs (14 points)
Bonus points for STEM emphasis (15 points)
Developing and implementing common, high-quality assessments (10 points)
Intervening in the lowest-achieving schools and LEAs (10 points)
Making education funding a priority (10 points)
Accessing and using State data (5 points)
Demonstrating other significant reform conditions (5 points)

You can see that the A #1 priority was to generate a good batch of paperwork and a well-polished application, because paperwork is the lubricant that greases the wheels of government. Close behind is a tautological statement (meaning "improve teacher performance based on performance" because silly bureaucrats are silly) that presumably means "get some kind of VAM system in place." The teacher effectiveness piece was worth more than the bottom six items put together; teacher effectiveness plus good paperwork is about equal to the last ten items on the list. Both are broken down in further detail in the document; the performance criteria boils down to "develop an evaluation system based on student test scores and use it to make personnel decisions (including "removing ineffective tenured and untenured teachers").

Next we get the other linchpins of RttT/waivers-- common core, charters, and turning around failing schools. Note that "turnaround" is worth four times as many points as "intervening." Making the funding for sustaining all of this a priority (which means, what, we promise to make serious faces about it-- how does anyone measure this) comes in close to the bottom. I was surprised that developing and implementing testing comes in close to the bottom, but I suppose you can just buy those, and their necessity is implied by all the top items, anyway.

The "move good teachers around" item (you know-- the one that absolutely nobody has figured out a plan for yet) comes in the middle of the pack. One does wonder what bureaucratic fiddling resulted in making that just one point more important than a statewide longitudinal system. How fascinating must that conversation have been?

These criteria are also arranged into six groups, and if we put those in point order, we get

Great Teachers and Leaders (138 points)
State Success Factors (125 points)
Standards and Assessments (70 points)
General Selection Criteria (55 points)
Turning Around the Lowest Achieving Schools (50 points)
Data Systems to Support Instruction (47 points)

But wait! The document also mentions the Big Priorities, which, as it turns out, don't entirely match the point system listed above, but which are ranked with fun language.

Priority 1:An "absolute" priority. The plan must be comprehensive and LEAs have to be all in.
Priority 2: "Competitive" This is the STEM bonus points.
Priority 3: "Invitational" Improve early learning outcomes. Which means you have to measure them, which means say howdy to standardized tests for four-year-olds
Priority 4: "Invitational" Ramp up your data system. In particular, make it connectable with other states' systems.
Priority 5: "Invitational" P-20 alignment, aka cradle-to-career pipeline.
Priority 6: "Invitational" Allowing school-level reforminess including control of staffing, budget, class alternatives, etc.

There's also a fun glossary of terms which didn't tell us anything new, but are good reminders of what the Department means. For instance, let's chase this definition-

Effective teacher means a teacher whose students achieve acceptable rates (e.g., at least one grade level in an academic year) of student growth (as defined in this notice). States, LEAs, or schools must include multiple measures, provided that teacher effectiveness is evaluated, in significant part, by student growth (as defined in this notice).

Student growth means the change in student achievement (as defined in this notice) for an individual student between two or more points in time. A State may also include other measures that are rigorous and comparable across classrooms.

Student achievement means —
        (a) For tested grades and subjects: (1) a student’s score on the State’s assessments under the ESEA; and, as appropriate, (2) other measures of student learning, such as those described in paragraph (b) of this definition, provided they are rigorous and comparable across classrooms.
         (b) For non-tested grades and subjects: alternative measures of student learning and performance such as student scores on pre-tests and end-of-course tests;student performance on English language proficiency assessments; and other measures of student achievement that are rigorous and comparable across classrooms.

Yes, when lost in the haze of debate and discussion, sometimes it's best to go back to the basics. Here it is-- exactly what the feds wanted. Good paperwork. A teacher rank and rate system based on student test scores that would drive everything from training. More charters. More school takeovers.

While the document says that RttT "will reward states that have demonstrated success in raising student achievement, that's not really what it rewards. It rewards states for remaking their education systems along the lines demanded by the feds. And though the document promised that the best models would spread their reform ideas across the country, five years later, there are no signs of any such spreading infection. But then, there are no signs that any of these federal ideas about fixing schools has actually improved education for any students in this country.

If Congress actually manages to shut this mess down, there will be no cause for tears.

Wednesday, December 17, 2014

Who Measures the Rulers?

Nobody squawked much when it was announced that Pearson had won the bid to develop the framework for the 2018 PISA test. The PISA, you will recall, is administered by the Organization for Economic Co-operative and Development every three years, leading directly to a festival of handwringing and pearl-clutching as various politicians and bureaucrats scramble to squeeze statistical blood from the big fat turnip of test results.

And yes, Pearson just won the right to design the 2018 edition. Given that back in 2011 Pearson won the contract to develop the 2015 PISA, the new contract is not a shocker. Given that Pearson is marching toward becoming the Corporation In Control Of Universal Testing, this barely qualifies as a blip. They have the GED. They have the PARCC. They have dreams of managing via computer every test, testlet, and testicle that exists.

There are many problems with that, but one of the fundamental issues is the one raised by this post's title.

When one person with one ruler does all the measuring, how are we to know if he's correct?

If we want to confirm the accuracy of our Pearson measuring tool so we check it against our Pearson standards device and make sure those results line up with the Pearson Master Assessment-- well, at the end of all that, what do we really know?

If Pearson tells us that our six-inch long baby pig weighs 500 pounds, how are we to discover that it's a lie? If Pearson weighs our bag of gold and tells us it's worth $1.98, and they own all the scales, how do we know if we're being cheated?

It doesn't matter whether the people who make the rulers are devious or incompetent-- if there is no one left to check their work, how do we know the true dimensions of anything? If Pearson makes all the tests and keeps assuring us, "Yessiree, this test lines up with our other test and fits in with the main test, so we can assure you that this absolutely measures true learning or complete education or intelligence or character or what matters in a human brain or the strength of a nation's education program," how do we check to prove whether that is true or not?

Who watches the watchmen? Who measures the rulers? To whom does Pearson answer, other than stockholders? I'm hoping we don't wake up some morning to discover the answer is "nobody."

Did Duncan Corrupt Common Core?

One of the recurring narratives among conservative supporters of the Common Core is the Tale of How Duncan and Obama Corrupted the Good and Virtuous Common Core. And no matter how often the tale is debunked, it keeps popping up again.

You can see the tale on display once again at the November gathering of GOP governors. The story always goes something like this:

Once upon a time, the governors (and some of their finest minions) got together and created a set of wonderful, magical standards. But just as they were starting to send these magical standards throughout the land, the Evil Presiden Obama and the wicked Secretary Duncan cast a terrible spell on the beautiful, healthful standards and overnight, the standards grew toxic tests that had to be taught to because of top-down federal intrusion.

I do not know if reformsters don't understand the implications of their own program or if they are purposefully deceptive (I'd guess there are some of each in play). But that fairy tales is not true, and never was.

First of all, there is no version of reality in which the states adopted CCSS on their own. Certainly there's no version of reality in which states would have adoptred the Core sight unseen without the federal leverage escape from the penalties of No Child Left Behind. CCSS fans can complain about feds, but it's the equivalent of complaining about the French-- we may not like them now, but nothing would have gotten off the ground without them.

But let's go back and look at Benchmarking for Success, the position paper for the National Governor's Association and their friends at Achieve. The document is no secret, and is often used to make the same point I'm about to make, but it's worth trotting out again every few months. And it's important remember that this report is from 2008. 2008.
ist2_4332306_new_year_2008.jpg
In their roadmap for education reform, Common Core is just one feature of the five recommended actions

1) Upgrade state standards to a common core
2) Use state influence to get textbooks, curricula, and assessments aligned to standards
3) Revise state policies on teacher prep, development and support
4) Accountability for schools and systems
5) Measure state-level achievement

One might look at this list and conclude that to accomplish this sort of large-scale overhaul would require a central planning body with a national reach and the power to back it up. So, you know, something like a federal government. But the authors of the "report" have anticipated that concern, and devote an entire page (well, two, but one is a full page picture of a brown frightened child).

If benchmarking were only about measuring and comparing outcomes, the federal government might be able to play a leading role. However, because benchmarking is also-- and most critically-- about improving policy, states must take the lead.

The authors assert that the states have the "primary authority" over the policy areas that are targetted by the reform. This is not an argument that we need to respect state autonomy; it's an argument that the state's authority stand in the way of the goals. When they write "the states must take the lead," that's not a philosophical imperative-- it's a recognition of a political reality.

They also note that the federal government can help by "playing an enabling role grounded in a new vision for the historic state-federal partnership in education." But the relationship is historic precisely because the feds are in it. The true historic relationship between the feds and the states when it comes to education is no relationship at all.

Nobody connected with this report is arguing, "We must initiate this great reform and keep the feds out of it."

Their specific to-do list for the feds is also not-very-hands-off. The feds should offer funding. They should do research and development. They should help identify the best benchmarks for states to use. They should collect and disseminate assessment materials. In other words, the feds should figure out the right thing to do, the right way to measure it, and decide who should and shouldn't get money.
Furthermore, the feds should "offer a ranged of tiered incentives" and those should include "flexibility in meeting requirements of existing federal education laws." In other words, the federal government should offer deserving states a way around No Child Left Behind.

In short, the federal government should hold the purse string of reform, oversee the definition of "deserving" for reform, and use the penalties of NCLB as leverage. They want the feds to send "support" for reform much like we once sent "advisers" to Vietnam.

Remember-- this report is from 2008. Do you remember who was not President in 2008? The same man who hadn't yet named Arne Duncan Secretary of Education.

Conservatives (and others) can argue that Common Core-related reform is tied to a large and unprecendented extension of federal authority. What they can't argue is that such overreach was the invention or creation of Obama and Duncan. Supporters of the Core got exactly what they asked for, hoped for, and planned for.

The most sobering part of these looks back is not the selective amnesia and political maneuvering among current conservative. It's the realization that the current reformster road map was in place before we even had Presidential candidates, which in turn makes me realize that the 2008 election was probably not going to have any effect on the future of US public education. The big question? Will the election in 2016 make any difference?

Originally posted at View from the Cheap Seats

Toxic Expectations

"Expectations" seems to be having a moment, shouldering aside "grit" in the pantheon of reformster Orwellian obfuscatory baloney.

To be clear, I am a big fan of educational expectations. I learned about expectations from my chemistry teacher Joe Stewart. We would whine that he expected too much and he would say, "I know. But if I expect this [hand held above head], I will get this [hand held at eye level]. If I only expect this [hand still at eyes], I will get this [hand at chest]." All of my experience as a teacher suggests that Joe had it right.

I communicate two things to my students with expectations-- 1) I intend to hold them to a certain standard and 2) I believe in their ability to succeed. We talk in my class about the pathway to awesome, not the road to good-enough-to-get-by. Students are fond of asking questions ("How long does this have to be? How much time should I spend on this?") that are basically reworded versions of "What's the bare minimum I can get away with on this?" My response is some version of "You are not trying to do the bare minimum. You are trying to be awesome. Don't settle. Be awesome."

Sometimes they want to offer some version of, "What do you expect from me? I'm dumb." They get in return, "I see no evidence of your alleged dumbosity. I expect you to be your version of awesome."
Okay, actually, I communicate three things. 3) I will be with you every step of the way. My role is support and guidance. On the trek up the educational mountain of excellence, I'm a sherpa. It's my job to egg them on. It's my job to make sure they have the supplies and support they need to get there. It's my job to gauge their strength and ability, to know when to say, "Come on! Let's go!" and when to say, "Let's set up camp and rest." It's my job to select a goal that will stretch them but not break them.

mountain.jpg
I tell you all this so that you know that I understand the power of expectations in education. But my understanding is apparently very different from that of many reformsters.

Arne Duncan has repeatedly insisted that students with disabilities are the victims of low expectations. The state of Washington, using Duncan's fact-free position paper as backup, insisted that

The evidence is clear that disabilities do not cause disparate outcomes, but that the system itself perpetuates limitations in expectations and false belief systems about who children with disabilities can be and how much they can achieve in their lifetime

Get that? All differences in outcomes are entirely the fault of the school. Students with learning disabilities, cognitive impairments, physical challenges-- they only have problems with school work because of the school. The blind student just has trouble seeing the PARCC questions because his teachers expect him to have trouble seeing. The student with limbs twisted by disease is unable to run a ten-minute mile because his phys ed teacher doesn't think he can do it. And the state of Florida was correct to demand that Ethan Rediske take the Big Test even if he was profoundly disabled and dying. We can expect that all children will be exactly the same, and we are just going to expect them all into magical compliance.

But it's not just that reformsters have imbued expectations with mystical magical qualities. Consider Erika Sanzi over at Education Post, Peter Cunningham's $12 million PR machine. She is ruminating on events and unrest in Ferguson, and about the question of the role of education in making young people feel valued:

When I think of how a school shows that it "values" children, my mind automatically goes to the question of expectations.

Does it? Does it really first go to the question of expectations? Because when I read that sentence, my mind automatically went to the question of baloney.

Do you know how a school shows that it values children? It does it by giving them just as much support as it can muster. It makes sure they have the best physical plant that money can buy, a school with all the amenities, a nice library, a well-equipped gym, classrooms that are clean and well-lit and filled with the best new resources that can be found. It spends top dollar to get the best people in the classroom. It makes sure the school can provide every kind of support, resource and facility possible. That's how a school shows it values children.

I am tired to the bone of reformsters claiming that expectations are all that we need, of the repeated chorus that we can't make schools better by throwing money at them. I have an experiment for testing that. Find the school in your state with the lowest level of spending, and reduce every single school-- including the schools in the wealthy neighborhoods-- to that lowest level of spending. When parents squawk, tell them it's okay because you are just going to load the expectations on. You are going to expectation the living daylights out of those kids and nobody is going to miss a cent of the money that was just cut, because, expectations. Try that, and get back to me.

You cannot truly deliver the expectation of success without becoming a partner in that success. You cannot help people climb the mountain without climbing it with them.

If you stand at the foot of the mountain and tell someone, "Get to the peak. Do it. I'm not going to check you out to see how high you can safely climb in one day. I'm not going to give you supplies or support, and I'm not going to help you, either. I'm just going to stay down here and expect that you'll make it to the peak, or else," that's not high expectations. That's just cruelty.

And to say to our poorest schools and communities, "You don't need to have the same kind of money and support and resources that the rich schools get. You just need expectations" is the lamest, most ethically lazy excuse since Cain said, "Brother? Um, where? Wasn't my day to watch him."

There is a half-truth in the reformster argument-- it is deeply wrong to look at students and say that because they are poor or challenged, there's no point in even trying. But it does not follow that by saying we expect them to succeed exactly like anyone else, we're doing any better. There are two groups that are ignored in this touting of high expectations: the children who have been rescued from low expectations by readily available money and resources, and the children whose high expectations have been crushed by the poverty and societal neglect that surrounds them. Neither group is aided by expectation blather, but only one group needs additional support.

When parents discover their child has a gift, they do everything they can to support it. Lessons, equipment, trainers, teachers-- even if they have to squeeze the family budget. They don't say, "That's nice, child. We expect you to be awesome, but you should not expect us to help you." There isn't an elite private school on the planet that says, "We have classes in a moldy, rat-infested barn. There will never be a nurse here when your child is sick or a counselor here when your child is troubled. We will do nothing special to assist your child whatever her difficulties. We have no books, no computer, and no facilities outside of the crumbling classrooms. But we will have really high expectations of your child, so send him to us!"

Nor does the parable say that the Samaritan found the man beaten and lying at the side of the road and said, "I'm going to do you a huge favor. I'm going to expect you to heal yourself and get yourself out of that ditch. Good luck. I expect I'll see you later."

To say that children who face the obstacles of disability or poverty simply need someone to expect more-- that's wrong in too many ways. First, it assumes that they are incapable of having expectations of their own, that they are simply idling and aimless, waiting for someone to slap them awake with a cold bucket of expectations (provided by people who know better than they what their goals should be). And second, it ignores our obligation to provide support, assistance, guidance, and even company on the climb to the mountaintop.
 
Expectations without investment are just empty promises and deluded dreams. They are excuses, a way to shed responsibility, to say that we have no obligation to help clear a path-- we can just sit back and expect the travelers to break their own trail, without even checking to see if they even have the tools to do it. The best expectations help show the way and light the road, but the worst are toxic, not only failing to push back obstacles, but adding the additional roadblock of Not My Problem indifference. If our students living in poverty don't feel valued, I'm pretty sure that the low expectations of their teachers are not the most likely culprit.

Originally posted at View from the Cheap Seats

Tuesday, December 16, 2014

Duncan Chases Teachers Away

Arne Duncan's new policy initiative is a perfect example of the law of unintended consequences in action.

Duncan proposes that teacher prep programs be evaluated by looking at the test results of the students of the graduates of the college. If that seems like a twisted sentence, that's because it's a twisted program. We can make two early and easy predictions about what effects it will have.

The first is simple. It will mean the college education departments will cut spending on programs so that they can afford whatever administrative assistant has to be hired to spend all their time chasing the numbers necessary to make the report to the feds. Some bunch of adjunct professors are going to have their hours cut so that somebody else can spend his days wending through the labyrinthian process of tracking down alumni, then tracking down their scores.

The second is, well, also simple. We already know that the best predictor of good student test scores is family income. Every college education department that doesn't want to get spanked by the US Department of Education has to do one simple thing-- they must do everything in their power to keep their graduates from getting jobs in poor urban schools.

Urban school districts that have tried to foster good relationships with college ed programs will find that they can't get their calls returned. College ed departments will screen school districts carefully and be cautious about which job openings they pass along to their grads.

If one of Duncan's goals is to put great teachers in poor urban classrooms, he could not have better designed a policy to do the exact opposite. This new policy is just one more step in the process of labeling some schools and some districts as career-killers, schools to be avoided at all costs if you wish to devote your life to teaching. This new policy will just add one more voice to the conversation saying, "Whatever you do, don't get a job at Poor Kid High School."

This is good news for TFA (or at least, it would be if they weren't suffering recruiting woes of their own) because Duncan's policy will help create more artificial teacher shortages in poor urban schools. But it is nothing but bad news for the schools themselves, branded with big scarlet F's and surrounded by signs screaming, "Whatever you do, don't come here to teach!" It is one of Duncan's poorest policy choices yet.