Wednesday, November 13, 2019

NAEP Board Gets DeVosian Additions

Education Secretary Betsy DeVos just announced six new appointees (and one returnee) to the National Assessment Governing Board. These are the folks that oversee the National Assessment of Educational Progress aka NAEP aka the nation's report card aka that big standardized test that everyone tries to use to prove a point every year.

So who did we get?

Returnee Alberto Carvalho, plus Frank Edelblut, Eric Hanushek, Reginald McGregor, Martin West, Grover J. "Russ" Whitehurst, and Carey Wright.

It's the reformster-heavy kind of group you would expect from DeVos. If the Curmudgucation Institute had a graphics department, they would totally be working on some sort of Snow White and the  Seven Dwarfs art.

Hanushek is an economist specializing in linking children's earnings to teacher effectiveness (I mean, what else could it be, right?) and who just recently mused that teacher pay should be linked to measures of teacher effectiveness, even though we don't have any.

Whitehurst is, for me, a mixed bag. A Brookings guy (boo) who once did a pretty good job of trying to get DeVos to say non-dumb things about choice (yay), but whose name is also on a "report" that argues we should test more (boo) and that when it comes to teacher evaluation, administrator observations should be more like VAM (booooo).

West is a professor at Harvard's uber-reformy GSE and editor-in-chief of Education Next, the Harvard GSE publication that carries work from the Fordham-AEI-Bellwether axis of reformsterdom. He's a Massachusetts state ed board member, too.

Wright is the chief state school officer for Mississippi, a state that is pretty awful when it comes to education, but they got their NAEP scores to go up, so she's an expert now. She's a Broadie, too.

McGregor is a "business representative" whose day job is the manager of engineering employee development and STEM outreach in the Research & Technology Strategy Group at Rolls-Royce Corporation in Indianapolis and he serves on a school board.

Edelblut is the real clunker here. Businessman, venture capitalist, and one-term state representative, then he ran for governor, lost in the primary and threw his weght behind Chris Sununu, who rewarded him with the education commissioner job. Edelblut has no actual education background; he didn't even serve on an education committee during his single term, and his kids are homeschooled. As commissioner, he has backed all manner of choice and vouchers and has happily signed up for as many DeVosian Freedom Bucks as he can get.

Oh, and Carvalho is the superintendent in Miami-Dade. You know DeVos isn't going to let a Florida guy go.

I have no idea how much mischief this crew can actually wreak, but it's not a group that I'm excited about having close to any educational policy activity. Add it to your list of reasons for not taking NAEP results so very seriously.


Pearson In Your Pants


Pearson, the edu-product giant that hopes to eat the world, just announced a new product.

It's part of the overall Pearson vision-- and nobody does large-scale vision like Pearson. They see everything happening in a "digital ocean." They have ideas about an "assessment renaissance" so huge that it took me five posts to write about it (here's the shorter version). And just this summer, they announced their intention to go "digital first." That is, not exactly phasing out textbooks entirely, but focusing on the digital; instead of offering digitized versions of print textbooks, they'll now work the other way around. Fun fact: "62% of Pearson revenue now comes from digital or digitally enabled products and services that make lifelong education possible."

It's very much in tune with their website slogan, "Learning Without Limits." That seems like a big reach, but again, Pearson has a big vision. What other textbook publishing company would offer two categories on its K-12 page: Products & Services, and Thought Leadership.
So this week's announcement is in tune with all of that.

Meet Aida Calculus. It's an app you can put on your phone, only that makes it sound too pedestrian, like one more version of Candy Crush. Hey, Pearson! Can we have some overly florid martket-speak here?

Aida is a first in the education industry and an important milestone in the use of artificial intelligence (AI) for learning. It's the only education app that uses multiple AI techniques from deep learning and reinforcement learning, customized for the purpose of tutoring students. The use of similar advanced AI algorithms is only seen in major consumer apps. Pearson is the first to apply that level of innovation in the education space.

Pearson CEO John Fallon boasts that "it's the first step we're making in redesigning education for the talent economy." It has thirty (30!!) "explainer videos." The Aida brand (which will be rolling out other edu-apps) takes its name from a combination of AI and Ada Lovelace (a 19th century mathematician and daughter of Lord Byron and just a hell of an interesting woman).

It can look at your homework, check your answers, offer tips. It uses AI that is the most AI-y AI that ever AI-ed in a piece of educational software. There's nothing AI-ier.

You can use the app for free for the rest of 2019, and if you want to try it and let me know how swell it is, I'd be glad to hear about it. I am in no position to try out a calculus app (my math education ground to a tortured halt somewhere around late-stage trigonometry).

Like all personalized [sic} learning, Aida offers several other implied promises, like someday you won't need a university or K-12 school-- you can just use your education voucher credits to buy apps for your phone that will automatically store your credentials on the blockchain and we'll never need any kind of formal education ever again. So I'm not going to encourage that by downloading it.

Will Aida actually work. Because ed tech folks have a history of overpromising everything all the time, with predictions and promotions that attempt to portray their product as an inevitable next step in human development. Maybe this time the super-duper AI will really deliver. Or not.

Actually, there's another reason I'm not trying out the app. Phone apps are like all those dumb games on Facebook (only worse) because if you're not careful, you are giving them al sorts of access to all the information about you that is stored on that phone. I'm not prepared to trust the folks who believe that the future belongs to those who have taken control of harvesting from the data ocean
(and who have already been hacked). Phone apps are convenient, a world of useful software right there in your pocket, but I'm not quite ready to let Pearson get into my pants just yet.


FTC Cracks Down On Edu-Influencers

One of the small tricks that education marketers have developed is to enlist teachers as brand ambassadors. Teachers are, after all, the voices most often trusted by other teachers, so it's got to be a real boost if you can get Mrs. Teachwell to tout your product on Twitter or Instagram or Pinterest. And the beauty of it is that Mrs. Teachwell may come cheap-- some free product, box of pens, maybe even some actual money.

It's a fuzzy ethical line; do teachers who accept some sort of marketing deal compromise their professional judgment? On the other hand, if you find an edu-product you really like, isn't the fact that you can get a few goodies for plugging just sort of gravy? People are not exactly lined up to give teachers things. Still, it's a pain to be talking to someone who you think is just offering a professional insight and then it turns out they're a paid-ish endorser.

Now the FTC has decide to add its voice to the conversation. EdWeek talked to FTC attorney Michael Ostheimer to get some clarity on the new rule, and it doesn't seem very hard tp grasp.

The connection between an endorser and a brand—whether it’s swag or a trip or getting paid money—that should be disclosed to the endorser’s audience,” said Ostheimer.

Just noting the connection in your profile is not enough. Every time you post about how awesome the Gradeinator 3000 is, you must also post that you are a Gradeinator 3000 Ambassador.  The guidelines indicate that you must reveal any "financial, employment, personal, or family relationship." Financial is of course not limited to being handed cash. Some of the guidelines are sensible (the notification has to be plain and clear and in the same language as the rest of the communication) and some of them will be  challenging. Liking a tweet or post by Gradeinator 3000 counts as an endorsement, so somehow your "like" will have to include your relationship. If you're doing a live stream that involves the Gradeinator 3000, you need to include your relationship frequently, so that your audience can't possibly miss it. (If you just really like the Gradeinator 3000 and you say nice things about for free, then you're off the hook.)

The rules don't seem to include second-hand benefits. In other words, my school district sent me to Seattle for PLC training, which was a definite benefit, but that was m district's money, not the PLC people's. But what about my SMART board training? The district paid for it, but the company certified me as a trainer, and my trainer status let me earn more hours toward my continuing education requirements, which were necessary for me to keep my job. That trainer status could have allowed me to make some extra side bucks as well, but I didn't. So did I get a financial benefit from the SMART board people or not? Ditto for anyone who has a collection of official Microsoft certifications.

The companies are supposed to get their "ambassadors" or "endorsers" or "influencers" or "good buddies" up to speed on all this, but it seems fair to be skeptical about that happening. If you want a quick primer, there's a nifty video that I'll embed. In the meantime, if you are all excited about the swag you're going to score by pushing the Gradeinator 3000, you might want to pause a moment and study up.






Sunday, November 10, 2019

ICYMI: Happy Birthday, Mom Edition (11/10)

Today is my mother's birthday, so there will be cake involved later. She's a pretty swell lady, so it's a day worthy of cake. Also, if you look to the right, you'll see that I've revamped the blog list and added a section of websites of interest that aren't necessarily blogs. So you can poke through that if you like. In the meantime, here's some reading from last week.

What Betsy DeVos Got Wrong About Detroit Schools

From the Detroit Free Press. Spoiler alert: almost everything.

Cassellius Puts Moratorium Onn District's Standardized Tests  

Boston's superintendent has concluded that an endless battery of practice and pre-tests might not be a great idea, so she's pressing pause.

The Failure of Betsy DeVos and 30 Years of Corporate Influence on Public Education

Nancy Bailey offers a brief but worthwhile history lesson about the true origins of public school problems.

The Stories of Segregation  Academies as Told By The White Students Who Attended

A fascinating new project gives us a chance to see an ugly chapter of US education history a bit more clearly.

Are Teachers Allowed To Think for Themselves

Steven Singer wants to know why teachers are highly trained, yet widely ignored.

More Testing Is Not The Answer For NYC Students, But Smaller Classes Could Be

Liat Olenick in the Gotham Gazette offers the cray thought that New York City schools might want to try a solution for which there is actual evidence of effectiveness.

What Is "Quality" Music? Choosing the Best Materials for Our Students

Nancy Flanagan is writing about the recent flap over what to include or throw out in the music ed library, but her thoughts here are useful for literature teachers, too.



Saturday, November 9, 2019

Digital Curriculum And Lesson Management Is A Crock

Ed Tech overpromises in so many areas, but one the great lies is that implementing This Year's Great New Program is going to save teachers just oodles of time. It never does. It particularly never does when it comes to the kind of software leviathan's used to manage curriculum and lesson plans. 

The sheer volume of data entry for these programs (enter the curriculum, unit plans, lesson plans, all cross-linked to the state standards) is enough to suck up hours of work on a weekly basis, particularly if your district has saddle you with one of the more user-unfriendly examples of the species.
 
"Well, yes, at first," is the standard reply. "But once you get everything entered, there's barely any work at all." Which is almost sort of true-ish if you teach mostly the same stuff in the same order on the same days from year to year. Sometimes you get the build-as-you-go argument-- if you enter all your units and all your lessons as you go this year, you'll just sort of build the curriculum organically. 

But it's still not true, because Ed Tech is just like regular old Tech when it comes to its love for the latest fashion. The tech world is not much interested in longevity and staying power: "Let's come up with a device with an effective life of at least a decade and power it with software that we'll never mess with but alway support," is not the rallying cry of any tech company. 

This blithe dismissal of the future is sometimes stunningly obtuse. Years ago, yearbook publishing companies were pushing digital yearbook supplements, extras that would carry lots of pictures and even videos stored on cool new CD-Roms. I'm sure that some schools went for this shiny deal, and now, over a decade later, those discs are useful as coasters or mini-frisbees, while old fashioned paper yearbooks from a century ago work just fine. 

But curriculum management software is arguably worse.

Consider the example of my old district. Our first attempt at digitizing the English department curriculum came over a decade ago. It involved some very user-unfriendly software in which we invested a mountain of person-hours until the administrator whose pet project it was left the district and the project was mothballed.

Around six or seven years ago, the district did a soft rollout of another package that was a little more sophisticated. Some of us were trained to be the vanguard, and then the rest of the staff was added. This was supposed to follow the cumulative model, with units and lessons building to a curriculum. An administrator decreed that all weekly lesson plans would be submitted on the software; their successor directed us not to worry about it too much. It was glitchy and unfriendly and some folks took the bet that nobody was actually paying attention, and stopped using it (some won the bet, and some lost). 

During one of the very first training sessions, somebody (I might have been the person) asked  just how long the district's commitment to this software would extend past the initial free-to-us grant-fueled period. 

The answer turns out to be, "Till right about now." There is, of course, no compatibility between software platforms, so my former colleagues are now apparently condemned to manually enter everything all over again. 

A fair question-- is all this accumulated and discarded hourage resulting in some beneficial curricular datafication?

Well. During the first round of digitizing, we had to do a lot of formating, phrasing and framing on our own, so I asked (I'm certain I was the asker on this one) what exactly was the purpose of the end product. Who was going to open this up, and once they opened it up, they would be using it to do...... what, exactly? I did not get an answer, though I did get one of those administrative Not A Team Player looks.

With Program #2, it was clear that one of the supposed benefits was a feature that allowed us to drag-and-drop individual state standards to each lesson, unit, etc, thereby allowing administrators to see what "gaps" we had in our standards alignment. This is a benefit if one assumes that A) having all the standards (created by a small group of edu-amateurs with no basis in research or evidence) represented makes curriculum better and B) that screen-weary teachers don't align their units by simply dragging and dropping every standard in sight just to get the job done. 

So, useful for administrators. But for teachers? For teachers, these programs are mostly a solution in search of a problem. Is it easier to write your lesson plan on a computer or a piece of paper? More importantly, is it easier to consult your lesson plan on a screen or on a piece of paper? 

Here's an underdiscussed ed tech issue-- paper makes for a really versatile display medium, while software usually forces you to look at your stuff the way the software manufacturer thinks you should. Lesson plans-- particularly if we're supposed to cross-check them with units and other weeks of instruction and state standards-- are a fairly complicated thing to display, particularly because it involves some information that the teacher may or may not need  at the moment. And then there's the issue of making on-the-spot alterations (e.g. didn't get all the way through exercise 12, so start with that tomorrow). It's entirely possible that there are teachers out there who start every lesson by pulling up the digital version of their lesson plan, but I haven't met them yet (you can speak up in the comments). Meanwhile, plenty of teachers still use paper to plan lessons, transcribe it into the software, and then use the paper to do the work (that, of course, in addition to teachers who don't use traditional lesson plans at all,  but that's another conversation).

There are other benefits to digitized curriculum. The more detailed standardized format makes it easier to, say, tell Mrs. McTeachalot to go pull up Mrs. Teachovik's planning for her Modern Numerology class and take it over.  If we could just get teachers to record their lesson planning in greater detail, we could more easily replace them, and if multiple teachers have the same course, you've got one centralized digitized record of exactly how they should all be teaching it. And if you use software that also allows for entering all the assessments, pretty soon you'll have your whole district recorded and transferable to any warm body. As an administrator, you can check up on what your teachers (say they) are doing. And did I mention you can check to make sure that all the standards are being covered. 

Yeah, you may have noticed that none of these benefits are actually benefits for teachers. I can think of one-- if the district decides to fund extra hours to get the data entry done, teachers have a chance to make some extra money doing work that will be thrown out in a few years. So much better than filling up binders that will gather dust on some shelf somewhere in the district office. 

Friday, November 8, 2019

FL: Even The State Thinks Florida Virtual School Needs To Shape Up

Virtual charter schools have a lousy track record, so bad that even bricks-and-mortar charter advocates have called for them to shape the hell up. Meanwhile, Florida has implemented every reformy method of undercutting public schools that could be imagined. So it's entirely predictable that Florida would have its own cyberschool, and that it would be a mess. But a lucrative one, as under previous governor Rick Scott, the state added a graduation requirement that every student must take an online course.

The school itself? Well, a 2014 Harvard Kennedy School study found that FLVS was "not significantly worse" than public school. For some reason, FLVS did not include that in their marketing.

Florida Virtual School was launched in 1997 by the state, and has since spun off into a sort of private beast that offers franchises to school districts. Problems have ensued regularly since. For instance, in 2018 the on-line school suffered a massive data breach,  characterized by one expert as not so much a hack as the school "left the door open."

The most recent round involved the Orlando Sentinel trying to plumb the depths of the work of Frank Kruppenbacher, the once-attorney of FLVS. The Sentinel learned that the lawyer had been mixing his FLVS work with that of other clients, jobs, and his own family. He used school employees to do work elsewhere, hired family, and, additionally, was apparently a sexist jerk to staff. The school kept all of this under wraps, including a nasty court battle; FLVS not only refused to release documents, but filed a civil lawsuit against the Sentinel for asking.

How bad did it get? So bad that state edu-honcho and former actively anti-public ed legislator Richard Corcoran recommended a big-time audit of FLVS, and Governor Ron "Never Met A Privatizing Scheme He Didn't Like" DeSantis agreed. Previous Governor Rick Scott had loaded the FLVS board of trustees with folks who were cozy with Kruppenbacher, and DeSantis has made some noise about changing that, too.

This summer a new FLVS head was appointed, and some board was shuffled as the state took control of FLVS. The new appointees are all politically connected, with nary a molecule of actual education experience in sight (though Corcoran does know one from church). So there's no reason to expect things to look up there.

And last week, the audit arrived, sort of. Giant chunks of the almost $200K report were cut-and-pasted from existing sources. Because, Florida. Nevertheless, the report recommends "a new governor-appointed board, new ethics standards for employees and a new inspector general inside the school to oversee internal audits and investigations."

It's another top-notch Florida-style swamp of shenanigans, and at this point the nations first state-wide cyberschool doesn't seem to have anyone in its corner. Though at the same time, the only solution being considered is putting someone else sit on the deck and watch as FLVS continues to wander adrift. Maybe someone should be suggesting to scrap the whole mess.




Thursday, November 7, 2019

Hanushek Offers Teachers A Grand Bargain

If there is anything we don't lack in the education sphere, it's economists who know all about how to make education work real betterer. Two faves are Raj Chetty and Eric Hanushek, who have both pushed some super-great ideas. You probably remember the one about how the right first grade teacher can mean you'll make umpty-zillion more dollars in your lifetime, a piece of foolery that might be called speculative economics, a house of cards based on so many tissue-thin assumptions, guesses and Hail Marys that it's hard to believe-- well, it boils down to sexy, memorable headline, so that's why you've heard it.

Anyway. Hanushek is over at EdWeek this week with a proposition, a proposal, a Grand Bargain, if you will-- but you probably shouldn't.

This guy, still.
Much of this is vintage Hanushek, but he's hung it on the current vogue for noticing that teacher4 pay is lousy and teacher working conditions are failing to attract hordes of awesome meat widgets. Sadly, this is what some folks think would be a grand solution.

Hanushek opens with some concern trolling-- teacher pay etc really is bad, as witnessed by tough job actions across the country over the past years, but "sequential appeasement of these outbreaks of union combativeness and teacher frustration will almost certainly not help the students and will likely make teachers worse off in the long run."

Hanushek is clear on the pay thing; he's done some of the research on the teacher penalty, aka the money that people give up to join teaching rather than other similarly-trained professions. And he is trying to dance on a thin line. On the one hand, he says, the low pay means that the teaching corps is not exactly the cream of the crop (and he throws in frickin' PISA results to underline his point, which is an odd  choice, given that US results have been mediocre forever, regardless of the ups and downs of the teaching pool). On the other hand, he says, "The bad balance between salaries and effectiveness does not mean that it is all right to bash teachers for not being better." And then he delivers his entry for the Backhanded Compliment Hall of Fame:

In fact, the nation ends up with a surprisingly good teaching force given the salary levels and working conditions. We attract many people who—for love of kids, for feeling of social purpose, or for what-have-you—are willing to take on the challenges of teaching.

Yes, I remember well the moment I realized that I could hear the call of what-have-you, and knew that I really wanted to pursue the something-or-other that could express my whatchamacallit somehow-or-other through kind of doing the teaching thingy. Thanks for noticing, sir.

But we're finally coming up on the meat of his argument, and it's the same old full-of-filler burger that he's pushed before. We need to fix schools by filling them with better teachers. How to do it?

First, he wants you to know that some teachers are more effective than others-- but we don't know what characteristics make some teachers more effective than others (but it's not education or longevity). Second, he wants to point out that simply raising salaries won't help, because it will retain both most and least effective teachers.

This is where the Grand Bargain comes in.

This bargain is simple: a substantial increase in teacher salaries combined with policies that produce a significant tilt toward more effective teachers.

Sigh. There are so many things wrong here, but this exercise in unicorn farming has been a favorite of modern reformsters.

First, we don't know what characteristics mark a highly effective teacher, so we inevitably go back to outcomes, aka test scores. The Grand Bargain has been proposed many times, and it's not that grand-- it just says "We'll pay the most money to the teachers with the highest test scores." But test score results are inconsistent and only cover math and reading. There's an underlying assumption here that's problematic-- the assumption that a teacher's effectiveness is some sort of set, permanent state, like their eye color or height. But with the exceptions of the extremes (superteachers and classroom disasters), most teachers are different levels of effectiveness on different days with different students. Teacher effectiveness is hard to measure in part because it is a moving target.

Hanushek has anticipated the argument about a lack of valid, reliable evaluation tools:

Mentioning evaluation often brings out a slew of arguments aimed at showing that any evaluation system—whether involving measures of student learning, supervisor and peer ratings, or parental input—has potential flaws. The claim that teachers can't be evaluated meaningfully stands in stark contrast, however, to what is seen in the vast majority of complex jobs across the economy.

Hanushek frames this as if these objections emerge as some rhetorical ploy; I'd say these objections emerge because there are real problems and people just keep pointing them out. And that last sentence-- I bet you think the next paragraph is going to offer an example of a complex job comparable to teaching that is meaningfully evaluated. Nope. He has no example to offer, and granted he has only so much space, but I'm going to argue that he has no example because there is no example because teaching is actually unlike any other profession. It stands in "stark contrast" to "what is seen" (by whom, anyway--oh, that passive voice) because the profession has some stark contrasts to other professions. For instance, health care professions might seem similar, but everyone who seeks medical help has the same basic goal-- to be made healthy. But the "customers" of schools have hundreds of different desired outcomes, from employable skills to bolstered self-esteem. Not to put too simple a spin on it, but education seems different because education is different.

So what else can Hanushek throw in? Perhaps some condescension.

To set a new, more positive path on evaluation, union leaders might take seriously one strand of their own rhetoric: We need to professionalize teaching. To some, professionalizing teachers means paying teachers the same as accountants. A more apt definition is professionals are people willing to be held responsible for their performance.

Really, dude? You are going to go with the old reformster whinge that teachers don't want to be held accountable for their performance? Because teachers are held accountable every day of their career. If you design a lousy lesson, you suffer some immediate accountability delivered by the roomful of small humans who will make you pay for your bad choice. Every parent has access to phones and email. Parenrts, administrators, board members, taxpayers--A teacher has a thousand bosses, and every one of them has some ideas about what that teacher should be doing.

Maybe what you want to say is that teachers should be subject to a formal accountability system that makes them pay a financial price for not meeting whatever standards we're measuring this week. But first-- as you've already acknowledged-- teachers have already paid a financial penalty for being teachers. And second-- and this is a huge one-- you don't have a functional, valid, reliable system for evaluating teachers, and until you do, teachers have to live with the possibility of those levers falling into the wrong hands, because here's the thing-- there is always someone who thinks we suck, someone who is a sure that we're a terrible teacher and a blot on the profession.

But to argue that because we don't want bad accountability systems, we don't want acountabillity at all-- that's just insulting. Teachers are just fine with accountability-- it's part of the what-have-you that drove us to teach in the first place, the drive to be able to look in the mirror and say, "You did good."

Hanushek winds up imagining that somehow such a system would involve union input and manages at the very last minute to throw in the old chestnut that "enhanced student achievement would engender broad economic gains across society." This is a piece of unsupported baloney,. but it fits the "if we just get everyone to score real high on the Big Standardized Test, poverty and income inequity will be erased and we won't have to address those issues at all" narrative. We've had years to see if this really works. Spoiler alert: it doesn't. One would hope that a truly professional economist would be willing to suffer some financial accountability for having pushed an inaccurate theory on education policy.