Thinking About Learning

Driving Simulators

2009-09-03T13:11:00.000-07:00

A recent NYT article about efforts to scare teens into not texting while driving has gotten me thinking about driving simulators. Again.

You may remember me talking a lot about these a few years ago. My premise, or chain of logic, went as follows:

The vast majority of driving does not involve crisis situations or need quick reflexes.
Thus, most drivers, and particularly most teen drivers, not only have no experience with driving emergencies but don't even understand how fast things go bad and how little time you have to respond.
Research suggests that kids who play twitchy video games actually make better drivers because they are more facile at tracking and reacting to many data inputs simultaneously without getting so focused on one that others get ignored. (I once asked former fighter pilot Rani Cohen what the most important characteristic is for fighter pilots, and he said it was without a doubt this same ability.)
Therefore, I reasoned, drivers could be trained on simulators, just like pilots.

I should add that some of this was driven by personal experience. When I was seventeen I wrapped a car around a tree after letting the tires drop off the edge of the pavement. Yes, I was told in driver's ed what to do, but I had never actually practiced it. When it happened to me my first reaction, my instinct, was to jerk the wheels back. Boom. Jaws o' Life, baby.

So how could I have actually been prepared for that moment? How could I have practiced? Simulators. The simulators I wanted to build would put drivers in one crisis situation after another, to give kids actual experience.

I did a lot of research, both into simulator technology and into teenage driving statistics. The technology situation was actually pretty encouraging; simulators are quite feasible and wouldn't have to be terribly expensive. But what I learned about teen drivers was less compelling. I found data that young drivers who take skid pad courses often become more dangerous drivers. Turns out that "skill" is only part of the problem; the bigger problem is that teens just think they're much better than they are, and giving them special training only exacerbates the problem. You can't just teach skill, you have to also teach wisdom. And how does one do that?

The most effective use of simulators, I concluded, would be that for every single student you turn the machine off after 5 minutes and tell them, "Sorry, but you're in the bottom 5% of all drivers, globally, and you're not even skilled enough to be taking this course. Go practice for a year...or ten...and come back." As long as the kids never figured out the scam it might work.

So I lost interest in simulators. (For a while I looked into developing a black box device that parents could hide in the family car that would tell them how aggressively their kid was driving. Not just how fast and where, but how hard they were accelerating, braking, and cornering.)

But that New York Times article has me thinking about simulators again, in a new way. The criticism of the shock approach is that kids already know traffic accidents are terrible, but that each one individually thinks, "Yeah, but it won't happen to me."

So here's my idea: build the simulator I was talking about earlier, but add a retina tracker so that you always know where the driver is looking. Now tell the drivers to try texting while driving. As soon as they're looking in their laps, that's when the kid runs across the road. Or the car ahead hits the breaks. Or somebody else (a virtual somebody else) makes a mistake like running a light or not signaling a turn.

The purpose wouldn't be to train their reaction times. In fact, it wouldn't be a game that you could "win" (except, perhaps, by never taking your eyes off the road). The purpose would instead be to demonstrate that if you look away at the wrong moment, no amount of reaction time and skill will save you. In other words, wisdom.

I hope somebody else builds this, because I'm busy with the Maven Project.

And on a related note....

2009-08-31T14:01:00.000-07:00

I've been too wrapped up in the Maven Project to take time for blog posts, but a couple of recent articles have gotten my economic rationalist hackles up.

Lately I've been emailing an article from The Atlantic, How American Health Care Killed My Father by David Goldhill, to practically everybody I know, so I figured I should probably just post it here. Even if you don't read this post, please go read the article.

I can't exaggerate my enthusiasm for this article. It's pretty much everything I've been arguing for years now, except argued more eloquently and with better data. The main point is that so much of our current health care debate seems to be based on the oversimplification that we have two mutually exclusive options, the free market or socialism, and that "the free market has failed in this case so it's time to let the government try." This article goes a long way toward explaining why the current morass is anything but a free market, and how the incentive structures for every category of participant...patient, physician, insurer...are utterly, completely FUBARed.

One point I wish the author had more emphatically addressed is the extent to which the "medical profession", insulated from real competition, resembles a guild or even a union, particularly in its adamant opposition to any sort of differentiating performance metrics or reporting. Just like the teachers' unions, the medical establishment wants to pretend that all of its members are equally competent, or at least that none of them are incompetent, that self-policing is perfectly sufficient, and that their behavior is motivated solely by customer welfare. So much so that despite their kicking and screaming about tort law, they'd rather deal with the roulette wheel of malpractice litigation than admit that mistakes are not, in fact, random or unavoidable.

As I said, I've been sharing this article around and a common response has been along the lines of, "Ok, a lot of that makes sense, but how are we going to fix it? We need to solve this problem now. So let's just go with the Obama plan first and then we can start fixing the incentive structures."

The problem with this solution is, of course, that not only will the problem just get worse due to expanded entitlements, but it will even further entrench the current interests. Incentive reform will be harder than ever.

On the the other hand, tearing apart our entire health care system and implementing David Goldhill's ideas is probably not feasible. Not only would it require even more dramatic change than the Obama Plan (which is primarily about health care finance reform, with vague promises about reforms in the actual provision that would happen once government bureaucrats had more authority; the Goldhill Plan would mean uprooting the entire system) but both insurers and providers would tend to be even more opposed to it. Although in theory a system based on rational incentives would be more profitable and more effective, large and established organizations typically don't like dramatic changes to their business models. In other words, the current regime may be utterly broken and illogical, but large hospitals and insurers have learned how to milk money from it, and they won't take kindly to attempts to change things.

The paradox in the Goldhill article is that after enumerating all the evils and unintended consequences of a system that is driven by mandate rather than incentive, the author then prescribes a set of sweeping changes that would come largely out of Washington.

A better approach, and one more philosophically inline with a true market approach, would be to plant some seeds of change and let them grow and evolve, eventually displacing the current structure.

Yes, I do have ideas for where and how those seeds might be planted, but it's going to have to wait for another post.

Union Busting

2009-07-22T09:33:00.001-07:00

I read this quote, referring to the repercussions of Arne Duncan's confrontational speech to the NEA that I mentioned previously:

In a town hall session after his speech, one NEA delegate said, "Quite frankly, merit pay is union busting," and got cheered by his colleagues.

My first thought is, "Damn straight it is."

However, it should be noted that the underlying purpose of merit pay isn't union busting; rather its purpose is to improve education. The union busting aspect is merely a side effect, due to the mutual incompatibility of merit pay (differentiated work) and unionization (collective negotiation).

If educational reformers also happen to take gleeful delight in crushing the unions, that's an understandable reaction to decades of stubborn union intransigence. If the unions can figure out how to add value in a pay-for-performance world, hey, that's great. Best of luck to them.

The tide is turning.

Where's our "War on Tenure"?

2009-07-03T13:04:00.000-07:00

Despite the historical ineffectiveness of declaring war on nouns, if we're going to have a War on Drugs and a War on Poverty and a War on Terror, we should definitely have a War on Tenure.

Yesterday Secretary of Education Duncan gave a speech in which he uttered the Truth That Must Not Be Spoken, i.e. that the NEA is intrinsically incompatible with the idea that some teachers are good and some are just bad.

"These policies were created over the past century to protect the rights of teachers, but they have produced an industrial, factory model of education that treats all teachers like interchangeable widgets.

The amazing thing is that this speech was delivered to the union.
This is, of course, what I've been arguing both in this blog and elsewhere: union's form to defend workers who, because they are in undifferentiated industries, derive no job protection from superior performance. Once a powerful union is in place it will therefore resist any attempts to institute differentiation, because doing so would obviate the need for the union itself.

Duncan seems to think the union can be persuaded to partner in reform, but while the individual members may be swayed the union itself will never bend. To do so would mean it's own death.

I can't help but think of Lenin, who was opposed to trade unions because he feared that workers would be sufficiently appeased to lose interest in the real revolution and the overthrow of the capitalists. So maybe I have something in common with Lenin after all, because I oppose any halfway measures that might appease the parents enough to let the NEA off the hook.

The NEA has harmed this country more...I do not exaggerate here...than Al Quaeda. Do the math. McKinsey released a report in April titled The Economic Impact of the Achievement Gap in American Schools (PDF download) in which they concluded that the gross inequalities in our educational system produce an economic drag equal to a permanent recession. Who gets the blame for the achievement gap? Most of it falls on the NEA for exactly the reasons Duncan aired (for which, by the way, he was booed and hissed by the teachers.)

The NEA kills more Americans than Al Quaeda, too, albeit indirectly. How many young urban blacks and Hispanics die annually to violence and drugs because our schools fail them? How many Americans die each year to Al Quaeda?

I hope Duncan is sincere about picking this fight with the NEA. The fact that it's coming from a Democratic administration means that it could plausibly happen. I only hope Duncan is smart enough to realize there's no middle ground for the union, and that he's only holding out the olive branch to lull them into a false sense of security.

(One of the benefits of not working for a large, respectable financial firm is that I can write stuff like this.)

2009-06-29T11:04:00.000-07:00

Whitney Tilson's newsletter included a link to this article in the Seattle Times. Apparently recent budget constraints in Seattle meant that some teachers had to be let go, and Seattle parents are up in arms against the teachers' union because it was the newest teachers who were fired, not the worst ones. Some of the pithy quotes from angry parents:

"Wake up and see how union refusal to consider merit is damaging the profession and our kids"
"We want the best teachers, not the oldest, teaching our kids"
"Teacher unions are an anachronism"

I can't help but think of the quote from my last (brief) post, “In retrospect, all revolutions seem inevitable. Beforehand, all revolutions seem impossible.”
The union will always argue, disingenuously, "But how can we objectively measure merit?" No matter what metric you choose, it's easy to argue how factors beyond the individual teacher's control skew the results and thus the lawyers can tie up assessment forever. It always reminds me of how Richard Perle effectively stalled years of arms control efforts using the same techniques.
However, Tilson also linked another article in the Boston Globe by an angry New Jersey parent, who proposed:

"To fix the system, we need first to identify problem teachers. This shouldn't be left to the principals. Every school should seek specific feedback from every student's parents, every year. Don't wait for unsolicited complaints: For every gripe that reaches the principal, there are many unhappy students and parents who didn't want to take that step. Discount the extreme responses, if you like; but don't ignore criticisms that are repeated over and over."

This is essentially what I've been saying for years: we don't need to use complicated measurement schemes; just ask the parents. The parents know who deserves to be given bonuses and who deserves to be fired.
Of course, the most direct way to "ask" the parents is not through surveys but simply by letting them pick who teaches their kids.

Revolutions

2009-06-22T16:13:00.001-07:00

“In retrospect, all revolutions seem inevitable. Beforehand, all revolutions seem impossible.”

--Michael McFaul

On Attracting Great Talent

2009-05-08T10:02:00.001-07:00

David Brooks has an essay, The Harlem Miracle, in today's NYT about the success of the Harlem Children's Zone, a charter school program in NYC.

Definitely worth a read. My only question is whether the model is replicable, or if it is dependent upon the personality of its leader, Geoffrey Canada. Coincidentally, an article in this week's New Yorker, The Instigator, is about a highly successful charter school program in Los Angeles, the Green Dot schools. As this article makes clear, the success of these schools is largely the result of the outsized personality of the leader, Steve Barr.

My concern with holding up localized success stories as examples of how education can work is that it suggests that other schools and districts could do the same thing if they tried. But that doesn't explain how we can attract more people like Geoffrey Canada and Steve Barr (and Dave Levin and Jaime Escalante and many others). Steve Barr, for example, sacrificed everything for the first few years in order to follow his passion. You hear similar stories about all of the most successful reformers. Do we really think we can find enough of those sorts of altruists to run all of our schools, and teach all of our classes?

The irony is that similar stories of hard work and ambition are easy to find in for-profit industries. We hear of rock stars sleeping in their vans while touring, or business titans sinking every last cent into a string of failed businesses until hitting the one that was a successful, or even young lawyers working 100+ hour weeks, and it seems romantic...but not necessarily noble. We know their sacrifices were made not for others but for their own futures, with the hope of reaping eventual rewards. Perhaps we don't admire these overachievers the same way we do the genuine altruists, but there are orders of magnitude more of the former than the latter.

If we want smart, driven people who will do whatever it takes to make schools work we can either hope that thousands more Geoffrey Canadas and Steve Barrs magically appear, or we can try to attract a similar level of talent and determination by offering actual careers and the possibility (but not the guarantee) of real reward. It's no secret which option I favor.

Addendum: just came across this analysis which raises questions about the validity of the data that David Brooks found so exciting. "Lies, Damned Lies..." and all that, but still. Particularly interesting is the discrepancy between HCZ children's performance on state vs. national tests. If "teaching to the test" is to blame, what does that say about the validity and applicability of our tests if the skills taught are non-transferable? It's hard to blame school administrators for focusing on the metrics by which they are assessed and funded, but the implication is that these tests are only for the good of the schools, not the kids. Again, let market forces decide how schools are assessed, not committees of bureaucrats.

Fast Girls Redux

2009-05-05T10:06:00.000-07:00

Yesterday I wrote about a story that brought back memories of a coaching experience I treasure.

In thinking more about that New Yorker article, How David Beat Goliath, I realized that there's relevance to one of my favorite topics, distance learning. One of the author's points is that anytime The Establishment feels threatened by a challenger that changes the rules, they will respond by attempting to discredit the newcomer.

"Insurgents work harder than Goliath. But their other advantage is that they will do what is “socially horrifying”—they will challenge the conventions about how battles are supposed to be fought. All the things that distinguish the ideal basketball player are acts of skill and coördination. When the game becomes about effort over ability, it becomes unrecognizable—a shocking mixture of broken plays and flailing limbs and usually competent players panicking and throwing the ball out of bounds. You have to be outside the establishment—a foreigner new to the game or a skinny kid from New York at the end of the bench—to have the audacity to play it that way."

And:

"The price that the outsider pays for being so heedless of custom is, of course, the disapproval of the insider. Why did the Ivy League schools of the nineteen-twenties limit the admission of Jewish immigrants? Because they were the establishment and the Jews were the insurgents, scrambling and pressing and playing by immigrant rules that must have seemed to the Wasp élite of the time to be socially horrifying...Goliath does not simply dwarf David. He brings the full force of social convention against him; he has contempt for David."

This is exactly the reaction we should expect and do in fact find from "professional educators" toward distance learning. Distance learning doesn't play by the rules of traditional educational establishments: it's scrappy, low-budget, and often openly profit-driven. It encroaches on geographical turf; it rejects the sanctity of the classroom; it completely bypasses the hierarchy. Most damning of all, it works.

Although I'd like to accelerate the distance learning revolution, in the long term my evangelism won't really matter. The incumbent in this case is particularly incapable of adaptation, and the insurgents will ultimately win this one.

Fast Girls

2009-05-04T13:22:00.000-07:00

My college roommate Tom just sent me a link to an article in New Yorker, How David Beats Goliath. (Tom reads New Yorker? Who knew?) I assume one reason he sent it to me was because it brings back memories...

Our senior year at Wesleyan Tom and I ended up coaching the Middletown High School girls' J.V. rowing team. The program was not highly competitive, and for a girl to skip practice to go shopping was both common and accepted. Tom and I put a stop to that. We told our girls, "We're going to make you work hard, but you are going to win." But, like, Ranadive in the New Yorker story, we didn't take the drill sergeant approach, and never berated them. One day when the other crews canceled practice because of rain we took our girls out on the water anyway, much to their horror. Motoring along in our launch next to 9 miserable, wet, bedraggled girls we started serenading them...ineptly...to an improvised plural version of the song "My Girl". It wasn't long before the girls we're grinning and trying not to laugh...as their boat went faster and faster faster through the downpour.

We had a couple of unlikely rowers on the team. Jen was a slight, arty type who had never done sports. On our first team run I had to run next to her, convincing her to just make it "to that next telephone pole" then "to that next corner" until she made it all the way back to the boathouse. Jen weighed nothing, but she was excited about the improbability of our mission.

I can't remember exactly how we found Janice. She was an 8th grader who we promoted to high school J.V., mostly because she was tall. Janice always had a sour expression and loved to grumble and whine and make sarcastic comments. But we put her in the six seat and teased her and flattered her into pulling.

We taught our girls everything we could that we had learned from Will Scoggins, a somewhat mythical coach who spent two years at Wesleyan winning almost everything we entered. The girls ate it up, and they went fast. Being cocky college boys, we kept challenging the coach of the girls varsity team to a race, but she always refused.

On race days we had a couple of psychological tricks. First of all, it was customary among the crews, particularly the girls crews, to use a lot more than 8 rowers to carry the heavy shells down to the dock. It was common to see twice that many kids carrying one boat. "No way," we said. "You're each plenty strong to carry one eighth of a boat, so you're going to learn how to carry them correctly." Partly we thought it was good for character, but we also wanted their competitors to see them carrying their boat without help. We drilled into them that while carrying the boats, or doing ANYTHING pre-race, they look straight ahead, never look at other crews, and don't laugh or smile. Intimidation.

Then there was our secret weapon: Jen. She wasn't very strong, but on cue she could belch like a guy. Crew races are tense at the start...boats are drifting slowly downstream while coxswains and judges try to keep them lined up and straight. All eight rowers poised at the catch, knowing the race could start any second. Coxswains put their hands up to signal "not ready" while they have their rowers make minute adjustments, so hands constantly shoot up then down again. In rough water and wind this can go on interminably. Suddenly the official yells "ALL HANDS ARE DOWN! SIT READY! READY ALL! ROW!"

Except in our races, in between the "Ready All" and "Row"....Jen would belch. We told the others, "Expect it. Ignore it. Don't think about it. It's just part of our start. Don't think about anything except what's going on in your own seat, with your own oar." Focus, focus, focus. Everything that we spent all spring teaching to them. Sure enough, when Jen belched we saw at least half of the girls in other boats look over. "ROW!"

Our girls rocked. They crushed. They won by so many boat lengths that we couldn't even estimate how far it really was. "Horizon wins" they're called. I'd be biking along the shore next to them, cheering and laughing so hard I almost couldn't ride.

That spring Tom and I graduated and I moved to Washington, D.C. I got a letter in the mail from little Jen, who wrote "Crew wasn't the same without you guys, so I quit and joined the cross country running team. I could never have done it if it weren't for you and the day you made me run." (For the record, I didn't make her run, I encouraged her to.)

As for Janice, the 8th grader, she went on to the U.S. Junior National Rowing Team, and rowed on the varsity crew at the University of Wisconsin.

Best job I ever had.

Ward Churchill Letter

2009-05-01T14:23:00.000-07:00

Figured I should link a recent letter that I sent to Boulder's lefty alternative weekly (which for Boulder is really saying something) that they actually published. Quoted in full:

(Re: “The right verdict,” Uncensored, April 9.) Pamela White has apparently confused the First Amendment with the tenure system.

The question in the Ward Churchill case was never whether or not he has a Constitutional right to spout nonsense, but rather whether the university must continue employing him now that his lack of professional integrity has come to light. The complication is that the fuss over the former led to the revelations of the latter.

The tenure system, not the First Amendment, very explicitly protected Churchill’s job when his “little Eichmanns” essay surfaced, so the anti-Churchill contingent started looking for other reasons to fire him, which they found. It turns out that Mr. Churchill is, at least to some extent, a fraud. The tenure that shielded Mr. Churchill is a two-way contract: “You get guaranteed employment for life, with summers off, and in return you adhere to professional standards.” Mr. Churchill failed to adhere to those standards.

The only relevance of the Constitution is that Mr. Churchill may have been emboldened by it to draw attention and scrutiny to himself, which is generally an inadvisable strategy for con men.

Mr. Churchill’s defenders argue that he wouldn’t have gotten caught for being a liar if it weren’t for the right-wing witch hunt, and that therefore the evidence is inadmissible. This is a little like arguing that Al Capone’s tax evasion was discovered only because Elliot Ness was out to get him and that therefore Capone shouldn’t have gone to prison. Maybe in the Churchill case there are some Colorado officials who improperly used their influence to trigger an investigation, and if so perhaps those officials should suffer some consequences, but the catalyst for the investigation does not change its findings.

I wonder what Ms. White’s reaction would be if it were discovered that a senator improperly used influence to initiate tax audits of AIG executives. Perhaps she would argue that any irregularities be overlooked because “they probably wouldn’t have been audited if not for the anti-executive witch hunt.”

The unearthing of Churchill’s deceptions and his subsequent prosecution for dishonesty is not McCarthyism, as Ms. White suggests.

It is, in fact, the exact opposite: an insistence upon honesty and integrity. In America you may believe and say whatever you want, but if you sign a contract saying you won’t make stuff up, don’t.

Lessons from Detroit

2009-05-01T09:12:00.000-07:00

The latest news about Chrysler gives us food for thought about the effects of unionization. It's easy to blame the unions for the failings in both (many) industries, such as in a recent excellent New Yorker article, but rarely do we hear a deeper explanation for the toxicity of organized labor other than basic greed. It certainly seems short-sighted of unions to repeatedly choke off their sources of employment.

Unions not only thrive but are necessary when labor is undifferentiated, meaning that productivity differences between workers has no impact on the measured effectiveness of the organization. Example: toll booth attendants are undifferentiated because the metrics upon which they could be assessed, say cars/hour throughput or customer satisfaction, are not metrics upon which the organization itself, and the compensation of its leaders, is judged.

Assuming that all workers get raises and that new hires earn less than veteran employees, collective negotiation is therefore in the interest of undifferentiated workers, especially the most senior ones, because otherwise higher pay rates become a liability: there is no reason to retain senior workers when their experience has no impact on organizational objectives.

So far that's all standard stuff. But think about what this system of incentives means for the union: their survival is dependent upon continued non-differentiation among their members. Thus we would predict that unions will fight, viciously, to prevent any actions which might lead to differentiation, which is in fact exactly what we do find. Once we understand this dynamic it becomes an illuminating lens through which to read the news.

Take for example this recent story about a supposedly successful "merit pay" experiment:

"It's payout time for Flora Childres, principal at Winchester Elementary, and every single teacher, counselor, coach and librarian on her staff, thanks to a merit-pay system even the teachers union supports."

"Even the union" supported this pay-for-performance plan...because no single teacher or principle was singled out for recognition; the merit pay was distributed equally among all of them. Non-differentiation.

I also liked (if one can describe the sudden urge to put a fist through a wall as "liking") a recent Washington Post story about Barack Obama being confronted by a pro-union teacher during a town hall meeting, in which

"he could not get the teacher to answer when he asked whether in her 15 years on the job she has encountered colleagues who she would not want to teach her own children."

Can't admit that some teachers are better than others, now can we?

This intransigence has the obvious effect of limiting our ability to recognize and reward outstanding teachers, or even to implement systems (think of the toll booth attendant example) that either measure differences or reward administrators for anything based off of differences.

I've long argued that the one thing most broken about our educational system is the absence of a particular kind of differentiation: horizontal specialization. We have vertical specialization in that teachers focus on different subjects, ages, aptitudes, languages, etc. What we do not have is specialization by function: at the K-12 level we do not differentiate between lecturers, graders, tutors, and other sub-components of teaching. Every teacher is expected to perform every role within his or her classroom.

No differentiated, professional industry works this way. Lawyers, architects, salesmen, engineers, bankers, executives...they all have junior assistants, secretaries, and other helpers so that the best employees can focus exclusively on the highest value work.

It's utterly insane that the few truly gifted, brilliant teachers we have are forced to write their own materials and do their own grading. Those great teachers should be spending 100% of their time in the classroom, working with kids. Let the newer and/or less gifted teachers do the grunt work...and learn from the experts while they do it.

But we can't. The union will never allow it.

The most common recommendation to improve our schools is to pay teachers more. This might work in situations where there is a teacher shortage, because (presumably) more people would apply for the jobs. But in cases where the gating factor is teacher quality not teacher quantity, an increase in applicants will not improve outcomes if there are no mechanisms in place for filtering out the good from the bad. The experience of the Teach for America program, in which many of the recruits don't make it through their 2 year commitment, is telling. They knew beforehand what the pay was going to be, yet they quit anyway. Why? Because they hated the job.

If we want to attract bright, ambitious, motivated people to teaching jobs, why do we model the industry after factory jobs? What kind of college graduate attracted to medicine or law or technology would be motivated by a career in which the job description never changes year after year, no amount of talent or hard work can result in promotion or increased responsibility, and success can never bring any kind of individual recognition? Those are the reasons the teaching talent pool is so lackluster. Schools are not competing for workers with publishers and law firms and technology companies; instead they are competing with public transit systems and assembly lines and the Department of Motor Vehicles. We're getting far too many teachers who are content with job security and steady pay, not people who want to succeed or fail based on their own ability.

Unfortunately, it's not so easy to get ourselves out of this mess. We can't get rid of the union without reforming the structure of the industry, and the unions have so much power that they will prevent any real reform. Chicken and egg. This is one reason I'm so enthralled with distance learning: distance learning companies are evolving and maturing outside the sphere of union influence, but are selling their services into the public system, or selling directly to parents who are pulling their kids out of the public system altogether. I hate to say it, but the only way we're going to have real change is by tearing down the current structures and starting afresh. If unionized factory-schools are the dinosaurs of education, I'm betting that distance learning companies are like the small mammals that darted around at the feet of the dinosaurs. We'll see who is left standing when the asteroid hits.

As Easy as Falling Off a Bike

2009-04-22T08:44:00.000-07:00

Learning to write has at least one thing in common with learning to ride a bike: adults who teach it typically don't actually understand how they themselves do it.

The physics that enable us to balance bicycles does not, contrary to common belief, have anything to do with gyroscopes. The actual principle is counter-intuitive and involves "intentional falling." When you start to tip over you steer toward the direction you are falling; your forward momentum then pushes you the opposite direction, tipping you back up. All bicycle riding, whether going straight or turning, is simply the application of this principle.

If you don't believe me, try this experiment: go ride a bicycle though a puddle of water and when you are back on dry pavement ride straight for a short way then turn left. Go back and look at your tracks. You'll see that your front tire took an initial little swerve to the right, just before your turn. That little swerve caused you to start falling to the left, i.e. the direction you really want to turn, at which point you turned the wheel the other way and carved into the turn. And you probably didn't even know you did that.

Bicycle physics are easier to explain with drawings than with words, but I hope this at least illustrates my assertion about trying to teach what we don't know. Most people who try to teach kids how to ride a bike don't actually understand how it works, and we do the exact same thing teaching writing. Since we can't explain either of these skills, we tell kids to "just keep practicing" and hope they figure it out. Sure, we give them lots of encouragement, and we explain how pedals and brakes work, just like we explain what nouns and verbs are. And we know good writing and good riding when we see them. But the actual art remains a mystery.

Knowing the physics, of course, does not mean that a kid can just start riding. But the knowledge can help us devise more effective learning exercises. For example, knowing that forward momentum is the force that keeps us up, it follows that it's easier to balance if you're going faster. Thus, a good learning strategy is take the pedals off a bike and let a child coast down gentle, grassy hills. By giving them free forward momentum they can focus just on the controlled falling aspect without worrying about pedaling.

It turns out that good writing can be explained, too, and this knowledge can help us give kids more productive instruction than just telling them to write about their summer vacations. Last fall I was introduced to the founders of the National Literacy Coalition, a Colorado-based organization that does professional development for teachers. NLC has a program for reading, but what they are really focused on...and excited about...is writing. They invited me a while ago to attend one of their training seminars, and yesterday I finally took them up on the offer.

The NLC program is derived from a successful curriculum developed in Australia called the "genre" approach to writing, which is based on the premise that there are really only five types of written work: Narration, Description, Instruction, Explanation, and Persuasion. Genre proponents will argue that any piece of writing falls into one of these categories. (A piece may have more than one component, but it will still have one purpose. For example, we may use narration to support persuasion, but our motive is still to persuade, not to tell a story.)

For each of the five genres (Americans often say "author's purpose" instead of genre) there are specific techniques and principles for how to write effectively, which are introduced gradually and with increasing complexity. Although the curriculum is roughly organized to go from kindergarten through high school, NLC repeatedly stresses that stage is more important than age. In fact the first thing an NLC-trained teacher does is assess where each child is; older students with poor writing skills may have to start from the beginning to learn the basics, although fortunately older students typically master the earlier skills more quickly.

My involvement with literacy has previously mostly been on the reading side. For the past 6 years I've been advising the Urban Education Exchange, a NYC-based non-profit with the goal of narrowing the reading "achievement gap" between poor urban kids and their suburban peers, using an innovative curriculum that emphasizes inferential readings skills. I became a believer the day I saw 3rd graders in Harlem passionately debating the "meaning" of a story, including citing textual evidence to support their positions. (I don't think I did anything similar before high school.)

So until very recently I equated literacy with reading. Reading, reading, and more reading. It never occurred to me to ask about teaching writing, maybe because it seems so much more more of a challenge than reading, which in itself is daunting. What the NLC made me realize, though, is that genre writing and inferential reading are really two sides of the same coin, and if you can teach kids the techniques for writing clearly and persuasively, they're going to recognize those same techniques in their reading. I'm not suggesting that we only teach them to write and trust that the reading will happen as a by-product, but the two should be taught in synchrony.

I've been talking with NLC about the potential for moving their curriculum on-line, and we've come up with some exciting ideas for how technology could be applied. NLC's emphasis on "stage, not age" makes distance learning appealing as a delivery vehicle because it neatly avoids the discipline and stigma issues that usually come with mixed-age classes.

I'm looking forward to more conversations with NLC. Stay tuned for more news.

Risky Business

2009-04-18T09:15:00.000-07:00

"When you got nothing, you got nothing to lose." -- Bob Dylan
This week I attended a Silicon Flatirons conference titled "Putting the Entrepreneurial Ecosystem in Perspective". A recurrent theme was the question of what makes a good entrepreneur and, at a deeper level, what drives some people to be innovators and what role education and natural intelligence play.

The discussion brought to mind yet again a conversation I once had with Gene Fife on a long flight back from an emergency weekend meeting. Gene kindly asked me some questions about my own career and interests, then said (I'm paraphrasing from memory):

"You take a guy who goes to a top prep school and then an Ivy League college and then Harvard Business School, and most of the time he'll have a willingness to take risk bred right out of him. He knows that he's going to have a great career and make a lot of money on Wall Street or in consulting, as long as he doesn't screw up. But you take a guy who got his MBA in Arkansas or South Dakota and he can't get that great job on Wall Street. So if he wants to make serious money he's got to start his own company. That's how you get your real revolutionary change. That's how you get a WalMart"

Regardless of what you think of WalMart as paragon of innovation, the larger point is food for though. I doubt Gene realizes the impact our conversation, which I think about frequently, had on me.

Coincidentally (but really not) one of my favorite people, Mario Morino, posted yesterday a new essay about risk and entrepreneurship. Mario talks about the factors that drive innovation, and while the pressures he talks about are environmental rather than personal the net effect is the same: risks are reduced when you have less to lose:

"If necessity is the mother of invention, then this crisis, which has laid bare the depth of our needs, provides us the dramatic necessity to drive innovation and spur entrepreneurs of all types and sizes to find ways to deal with our challenges."

I hope Mario and Gene are both right. I hope the silver lining to the economic crisis is that as people are driven out of comfort zones they might not have willingly left they take risks they might not have otherwise considered. Considering all the innovation we've experienced during boom times it will be interesting to see what emerges out of the crisis.

On Leadership, Evolution, and Large Furry Mammals

2009-04-17T13:31:00.000-07:00

Judith Warner, whose columns I generally like even when I don't completely agree with her, has recently written "Dude, You've Got Problems" about adolescent boys and homophobia. I will resist quoting the entire column because you should go read it yourself, but the gist is that Warner and others think that we somehow need to teach boys to not use the word "gay" pejoratively and in general to not be so cruel to each other.

While I sympathize with the motivation, I suspect the goal is quixotic.

The essay has a couple references to how adolescent prejudices and predilections haven't changed in decades. "They were frozen in time," one of the researchers says. While this is offered as proof that society isn't doing enough to change things, it is actually an unintended hint at the root cause: the behaviors described are not cultural, they are genetic. Of course they are frozen.

You know the co-evolution story. The caveman who wins at mastodon wrestling gets all the girls. They have lots of grand-kids, among whom mastodon wrestling is a natural talent among the boys and an attractive quality to the girls. (Well, Alpha-ness in general is attractive, not mastodon wrestling specifically. Even today the subtleties of mastodon wrestling remain insufficiently appreciated by most women.)

The evolutionary result is that males are hyper-conscious of hierarchy and instinctively motivated to both claw their way up it and to only associate with (i.e., rely on) males of equivalent status. In some circumstances hierarchy may be determined partly or even mostly by earning power, but among cavemen and adolescent boys it's about physical and sexual power.

In the hierarchy game, bullying, ridicule, and picking on the weak is part of an evolved...and effective...strategy. Those who can't "take it" are inviting more abuse by triggering ancient instincts to drive weak members away. I'm not saying it's right, I'm saying it's how it is. (Too many observers seem resistant to evolutionary psychology...coughsharonbegleycough...because they're afraid that explaining the dark side of human nature is the same as justifying it.) Even if we convince boys to not call each other "faggot" we're not going to change their hard-wiring; they'll find some other outlet for it, as some of Judith Warner's readers commented.

By the way, it's worth noting that bullying as a strategy tends to be most effective for those who are not at the top of the hierarchy. Ed Massey's "Corporate Ladder" sculpture, while somewhat apropos, is misleading in that it's the CEO at the top of the ladder kicking the next guy off. Had the artist been making an evolutionary statement rather than a political one it would have been a middle manager. That point will resurface in the conclusion to this post.

When I was in boarding school "hazing" was quite common. I say that in quotes because it wasn't the kind of formalized ritual we associate (if only because of Hollywood) with college fraternity initiations; it was just boys figuring out the hierarchy, often in a rough way. Compared to what I endured growing up, in a neighborhood where I was two years younger than the next youngest kid, boarding school really wasn't so bad, and certainly (to me) was a lot less malicious.

Ten years later I returned briefly to my school as a teacher, and in the interim some particularly rough incidents...in my old dorm, in fact...had resulted in Hazing getting pushed to the top of the list of Things That Will Get You Kicked Out of Boarding School, above even drug use and plagiarism. Predictably, a side-effect of this War on Hazing was that every incident of kids being mean to each other, even basic name-calling, began to get characterized as hazing (kind of like how recently it seems like every crime gets called "terrorism"). This of course only served to debase the word.

The more troublesome but equally predictable side effect of the War on Hazing was that the real hazing went underground. The school tried to abolish any traditions, even informal ones, that seemed to condone any sort of hierarchy. Naturally the kids, particularly the boys, found other outlets for their instincts, but since their activities were now out of sight the "fun" started getting rougher. The faculty thought they were simply banning one type of activity among many, but they were really trying to legislate away human nature.

So my second year teaching I tried an experiment. In the Fall I held a meeting for all the seniors in my dorm. I told them, "I will never cut any of you slack or give you special privileges simply because you are seniors. Reaching the age of seventeen without getting kicked out of school is really not a very impressive feat. Nevertheless, the underclassmen inexplicably look up to you anyway, apparently for no better reason than you're older than they are. There are two things you can do with this completely un-earned power: you can be feared tyrants, or you can be respected leaders. Being a senior does not in itself impress me, but leadership does. If you want special privileges you can earn them, but they won't be handed to you on account of your age."

I'm paraphrasing, but this was the gist of the lecture I gave, and of the reward system I enforced. I was trying to channel their hierarchical instincts in positive ways. Although I won't claim that it was 100% successful, or even close to it, there was in fact a noticeable difference that year, including some gratifying incidents in which upperclassmen intervened in defense of younger students. As emotionally painful as it is to be the victim of bullying, there is no better salve than support and friendship from somebody who outranks...who has more status and is "cooler" than...the bully.

Getting back to the Judith Warner column, if we want to make adolescence more bearable for boys who are not the alphas there are more effective strategies we could pursue than trying to banish instinct. The final part of her essay gives a clue about how to do this:

[One researcher] found that parents played a key role in reinforcing the basest sort of gender stereotypes, at least where boys were concerned. “There were a few parents who were sort of alarmist about whether or not their children were going to be gay because of their music choices, the clothes they wore,” she said. Generally, she said, “there was a kind of low-level paranoia if these high-school-age boys weren’t yet seriously involved with a girl.”

It seems it all comes down, as do so many things for today’s parents, to status.

If we want to have a positive effect on adolescents we have to have the parents on board. As the above quote illustrates, a campaign to simply convince boys to be "nicer" would face genetic obstacles not just in the boys themselves but in their parents. Humans did not evolve to willingly sacrifice their own children's mating prospects for the good of the village; those who did so disappeared from the gene pool many, many generations ago.

Instead of trying to stamp out the hierarchy instinct, I think my boarding school experiment suggests a more feasible solution: find ways to focus that instinct in productive rather than destructive ways. Educate parents to realize that boys who act like leaders rather than tyrants are not weak themselves but are in fact increasing their own status. Teach moral leadership not because it is "nice" but because it is a path to success...and gets the girls.

The advantage of the leadership approach...aside from the part about not flying in the face of eons of evolution...is that it doesn't require much market penetration to be effective, unlike the sensitivity approach. Instead of futilely trying to suppress negative behavior in everybody, let's encourage positive behavior in some. We'll never eliminate all bullying and name calling, but we can mitigate the damage by cultivating leaders who will police and protect their own.

Poetry in Motion

2009-02-16T09:57:00.001-08:00

Ok, ok, ok....I know it's been a while since I've posted. I'm going to change that. Really.

Yesterday I learned to skate ski. At one point, while leaning on my poles gasping for oxygen (which seemed to comprise a large part of the learning process) I was reflecting on the phrase "poetry in motion" and it's metaphorical origins. It occurred to me that metaphors are typically the use of the familiar to describe the unfamiliar, but for many people today, particularly kids, the phrase "poetry in motion" is a backwards metaphor: thanks to the ubiquity of video we are more familiar with visual poetry than we are with the real thing. Standing there in the snow I thought that the phrase "linguistic athletics" might actually be a more illuminating metaphor: use sports to explain writing.

In fact, there might even be a practical application of that idea. Imagine showing some short video clips of amazing athletic feats to students in an English class, and/or asking them to find similar clips that they like on the Internet, then asking them questions such as these:

What exactly it is that makes these scenes so fun to watch.
What attributes are the athletes displaying that are exceptional? Speed? Power? Reflexes? Grace? Awareness?
To what extent does the outcome (a point scored, a tackle made, a stunt landed) affect the overall allure of the scene? In other words, is it what the athletes did, or is it how they did it? Or both?
How many times did you watch it? Why more than once?

It should be obvious where this is leading, but the next step is to introduce that idea of linguistic athleticism. That fine writing is not simply about what is being said, but in how it is being said, and that we can appreciate the skill of truly fine writing. If reading prose is like watching a whole game, reading poetry is like watching game highlights. And great poetry is like those few, rare video clips of truly exceptional moments.

Furthermore, good poetry isn't defined by just one characteristic. Some poets (Gerard Manley Hopkins) excel at the texture of language, how it actually sounds. Other poets (Billy Collins) amaze us with their surprising metaphors and twists. Others are masters of imagery (William Carlos Williams) or their ability to evoke emotions (Pablo Neruda) or the intricacy of the poem's very structure (Paul Muldoon). On other words, there are just as many ways for writers to amaze us as there are for athletes to do the same.

Homework assignment:

Pick a poem that you like.
Identify what it is about the poem that appeals to you.
Use the language of sports to describe the poet's talent.

I think it was about that point that I remembered I was still standing out in the snow, no longer gasping and now starting to shiver in the utter quiet. Poetry in stillness?

Craig Barrett on Education

2008-10-07T11:10:00.001-07:00

The other day the New York Times had a short piece with the catchy title: "Intel's Barrett: Teachers Matter More than PC's". The closing Barrett quote is:

PCs aren’t magic. Good teachers are.

The article also has a reference to a book by John Kao, "Innovation Nation" (subtitled "How America is Losing its Innovation Edge, Why It Matters and What We Can Do to Get It Back") with a quote on the same topic of teacher quality:

Finns have come to cherish gifted educators as Texans do ace quarterbacks. The country’s teacher training schools have 10 applicants for every opening.

I suppose the response of some people will be, "See? We have to pay teachers as much as quarterbacks!" Hmmm.

To repeat a theme on which I continually harp, remember that most aspiring quarterbacks don't make any money at all. Quarterbacks are under constant pressure to prove themselves against their peers, and only the very tiny few who end up being the best in the world become rich and famous.

Trigonometry 2.0

2008-10-01T11:06:00.000-07:00

It seems that every time I get into a conversation about education with somebody new I end up telling the same anecdote. It's about an experience that more than any other single factor led me into education.

My first real job out of college was at a large advertising agency but in small group that was developing some innovate software to support marketing. On the second day my new boss said "teach yourself to program so you understand what we're talking about" and handed me a book on Lisp programming. What I've realized since then is that my wiring is conducive to writing software, because soon thereafter I was writing code full time (and then some).

An early task was developing a graphical representation for some SAS statistical output called "correspondence analysis" that didn't lend itself to the standard palette of Excel charts. After I got the basics working I started thinking that what my chart really needed was a 3rd spatial dimension. But the toolkit I was working with only had functions for 2D graphics.

"How does one take a three-dimensional object and draw it realistically on a two-dimensional surface?" I wondered. It occurred to me that it would require a projection similar to a shadow, so I cut up some manila folders, taped the pieces together into some boxes and cubes, and held them underneath a desk lamp so I could observe the shadows as I rotated them.

"Hmmm," thought I, "as I rotate the cube ninety degrees, the projection of one face of the cube goes from 100% to 0%, while the adjacent face does the inverse. That reminds me of something I once did in 10th grade math....what was that called?"

Trigonometry. I stared at my paper cubes and tried to retrieve any trigonometry knowledge lurking in my brain, but the only memory that wafted to the surface was the mnemonic "soh cah toa". (Sine is Opposite over Hypoteneuse, etc.) I borrowed a scientific calculator and started playing around with numbers ("Oh, right....radians not degrees. Hmm...what's a radian again?")

As I reconstructed my lost trigonometry knowledge I started encoding my discoveries as software. At the end of the week I had a red cube slowly rotating on my screen. I went on to add realistic shading and perspective, and eventually created my own library of 3D graphics routines.

As I watched that red cube spinning I reflected on 3 things:

I was immensely, deeply satisfied. I couldn't get enough of watching my little program do its thing, with the thought "I made that."
Although I had worked extremely hard getting to that point, working late into the night and obsessing about the puzzle while commuting or laying in bed, I had fun the whole time.
I knew I'd never "forget" trigonometry again. In fact, because of that experience I understood the essence of trig so well that for a long time afterward I would think about sines and cosines while watching a door swing shut or the wheel of a truck turn. Trigonometric relationships, I discovered, are beautiful.

Given those three discoveries, there is one obvious question: Why don't we teach trigonometry this way in school?

I'm still trying to answer this question.

Career Trajectories

2008-08-24T11:45:00.000-07:00

In my post about the book Disrupting Class I concluded with three questions, two of which I've addressed in subsequent posts (here and here), leaving one final one: is there any pragmatic hope for technology to overcome the political and cultural obstacles to such a dramatic change?

The book's answer is an extension of the theme started in The Innovator's Dilemma and is itself convincing, although limited in scope, a point I will address shortly. Their argument goes as follows:

Disruptive technologies wriggle into change-resistant systems by addressing the needs of under-served audiences, largely because those markets are under the radar of the dominant players.
In the "business of education" those under-served markets exist in the margins of budget-constrained schools: the AP classes that don't exist in rural schools; the science and language courses being squeezed out of larger schools by the pressures of NCLB; the specific structure needed for the disruptive ADHD kid.
Although our distance and online learning technologies are still nascent and imperfect, they still are (or can be) better than nothing, and are thus attractive solutions for these needs.
The experience of serving these difficult and fragmented markets will help both refine the products and to improve public perception/acceptance, allowing the new technologies to move incrementally upstream.
Eventually the effectiveness of the ever-improving technology is demonstrated sufficiently to begin directly challenging the incumbents, for whom it is by now too late to adapt.

This model works fine for explaining how a disruptive technology can supplant an older one, which in this case would mean Internet-based instruction replacing traditional live classroom instruction, but it says little about the workers. That's sufficient if we are talking about radios or hard drives or even self-guided online courses where we're focusing on the interaction between customers and technologies. If we think teachers will get replaced by software then we've answered the question. But if we agree with my previous hypothesis that technology will change the job of teaching without replacing it then there's still something to explore.

Let me put this another way: given the problem that "not enough of our best and brightest want to be teachers" some people might think technology is a compensating solution in that it shifts the burden of quality to replicable automation. On the contrary, I think technology can be a catalyzing solution in that it will make the profession more attractive to talented people. I base this prediction (or hope?) not solely...or even primarily...on the assumption that technology will directly change the mechanics of teaching, but on the premise that new companies delivering online learning services will be free from the rigidity of our calcified, 19th century educational system and will therefore be able to offer not just jobs but careers that smart, ambitious, talented people will find appealing.

I hear over and over again that the "problem" with our schools is that we don't pay teachers enough, and if we paid them more we'd get better teachers and better results. I don't think this is exactly true for two reasons. The first reason is that higher pay does not in fact insure higher quality, unless there also exists a filter for that quality, i.e. both a reliable performance metric and a mechanism for culling low performers. Otherwise you have no way of figuring out which among the expanded pool of applicants are the good ones. (Note: I actually just typed an unreasonably long expansion on this argument, but I'm going to defer that argument to a future post, partly because I'm going to have to draw some graphs of a type that Excel can't handle.)

The second reason, and the real topic of this post, is that I don't believe higher salaries in themselves are sufficient to attract our most talented people. Look at all the passionate, talented people that work for a pittance as Capitol Hill interns, or take underpaid jobs doing menial work in publishing houses, or deliver pizzas to subsidize their rock bands, or wait tables while writing/acting/singing. These careers often pay less than teaching, yet still attract motivated, determined young people. Why is that?

I think it's because of three interrelated factors:

Although the starting pay is miserable, the few who defy the odds and succeed ultimately get rewarded for it. And since we each think we're the hero/heroine of our story we tend to think those odds don't apply to us. So we go for it anyway.
As one achieves success in those careers the job itself becomes more interesting and rewarding. Performers play bigger and more important venues. Those in politics start at the photocopier and end up writing legislation. Writers start with short stories in obscure 'zines and eventually (maybe) end up on the Times' best seller list.
Finally, the ladder of rewards in those careers, monetary and otherwise, is determined by ability and effort and is not simply a factor of time spent on the job.

Smart, ambitious, driven people like to compete and want their efforts and success to differentiate them. Our "best and brightest" will never be satisfied doing the same thing year after year, with the same responsibilities as their less able peers, with their monetary rewards inching up at a predetermined rate. The counter-argument I often hear is that the system we have attracts people who love teaching for its own sake, i.e. the altruists, and that's a Good Thing. My response is always two-part: a) Often, regardless of the financial rewards, it's those who genuinely love a profession that ultimately are the most successful. b) In any event, do we really care what somebody's motives are, as long as they're good at what they do?

(I was debating this point with a close friend in graduate school who said, "But in your world we'd get mercenaries teaching, and I don't want mercenaries teaching my kids." I replied, "Ok, who would you rather have teaching them: the teachers you had in high school in L.A., or me?" "Well, you..." she said. "Right, and I'm a mercenary.")

Evidence for this view is that recruiting good people isn't the problem as much as retaining them. Take for example Teach for America, which is very successful at recruiting bright graduates from top colleges, but less successful at keeping them in teaching careers. Everybody knows that teachers don't make a lot of money, so if salary was the problem they wouldn't have started teaching in the first place. It could be that a lot of new teachers find out they just don't like teaching, but former teachers I've talked to (and this goes for me as well) light up when they talk about working with the kids. No; it is the career itself that is unsatisfying and unfulfilling.

Note also that all of these various points are interrelated: you can't reward people for performance unless you have a way to measure it. So we can't, in our present system, change these characteristics one at a time. The unions are an easy target for derision, but in the absence of a reliable performance metric a union is in fact necessary: systematically undifferentiated workers have to band together. If performance won't provide job security, collective negotiation has to.

So what I find so encouraging about distance learning, the kind that includes teachers, is that it offers a chance to build a new system from the ground up. The companies delivering these services will need not just technology but teachers, and they're going to have to compete not just with each other but with our existing schools to get them. But, free from the constraints of the legacy system, they'll be able to offer compelling benefits that are essentially absent in the present teaching profession:

The chance to work with a particular type of student drawn from all geographies, not just anybody of a certain age who happens to leave near a physical school.
The opportunity to be entrepreneurial, to experiment with different models and even to actively recruit one's students.
Rewards that directly correlate to one's ability to attract pupils.
Specialization in very specific areas and roles.
For proven, experienced teachers, the possibility of support staff (or technology) to offload the easy work, leveraging the teachers' talents across more students.
Effective and ongoing mentoring/training programs.

In particular I want to comment on #5. In The Mythical Man Month Fred Brooks explains why twice as many software engineers can't write software in half the time: the overhead of communication in a group scales exponentially, not linearly, as group size increases. His solution? Instead of throwing more engineers at a project, give your most productive engineers support staff to handle the low level work, enabling them to be more productive. I.e., leverage their talents more productively. This is exactly what we need in education. Instead of assuming that each teacher can only handle X students, we should identify our best teachers and give them the resources to reach more students.

The problem with not leveraging teachers beyond X students is that it negates a large part of the premise and power of a voucher-like systems. What's the point of creating a "market" for education if there is no economy of scale? A dynamic principal with a staff of the 10 best teachers in the district may have every parent in town banging on his door, but without leverage he still won't be able to admit more than 10X students.

A relevant anecdote: Just recently an acquaintance was telling me about his days selling cellular contracts in Oregon. He was pretty successful at it, but he had a colleague whose consistent contracts/week numbers defied logic. "It takes a certain amount of time to do the paperwork, handle provisioning, and do all the other steps required for each contract. I just didn't understand how she had time to get that many contracts done." Finally at a company event he had a chance to meet her, where he discovered two things: First, that she was beautiful and charismatic, which never hurts in sales. Second, and more interestingly, she told him that she hired an assistant, with her own money, to do the paperwork and provisioning.

That's the power of leverage. And that's what's missing in education. What I found most interesting about that story is that although the company didn't recognize and provide for leverage, it did at least provide the financial incentives, so one extremely talented and smart saleswoman who did the math found and paid for her own assistant.

I see two possible scenarios for distance learning companies. One is a platform in which a company provides technology support for "educational entrepreneurs" who want to create their own curricula and recruit their own students, possibly competing with other such entrepreneurs on the same platform. The other model is more like a school in which the company recruits both teachers and students and decides where to allocate resources, including support staff. In the former model the company takes a cut of revenue, while in the latter model it gives a cut. Either way the incentive of increased revenue drives the quest for leverage and efficiency.

In some ways I've strayed far off from the question about the ability of our culture to adapt to a new educational model. But the most important facet to consider is not simply how we as consumers react to the new technologies, but how teachers (and possible future teachers) will respond to the changes. My hope is that distance learning will end up attracting the greatest talent, much of which currently chooses not to teach, because the underlying business model enables careers that are so much more rewarding than the existing undifferentiated, unchanging, unrecognized alternative.

Teaching Science

2008-08-24T09:40:00.000-07:00

Today's New York Times has an article about a high school teacher in Florida who, like so many American teachers, has to contend with devout Christian resistance to theory of evolution. It's an interesting article about a gifted teacher, but what really caught my eye was the comment that:

Some come armed with “Ten questions to ask your biology teacher about evolution,” a document circulated on the Internet that highlights supposed weaknesses in evolutionary theory.

I was curious about this list and wondered what the questions were like so I Googled it, and the first result I found had both the questions and an informed answer to each one. I was expecting annoyingly simplistic yet tautologically unanswerable questions such as "If God is omnipotent and infallible, why would He intentionally make creatures that need improvement?" Instead the questions demonstrated intimate familiarity with the history of evolutionary theory, and unfortunately the answers required some appreciation for subtlety and nuance. An example:

DARWIN'S FINCHES. Why do textbooks claim that beak changes in Galapagos finches during a severe drought can explain the origin of species by natural selection -- even though the changes were reversed after the drought ended, and no net evolution occurred?

A: Textbooks present the finch data to illustrate natural selection: that populations change their physical features in response to changes in the environment. The finch studies carefully - exquisitely - documented how the physical features of an organism can affect its success in reproduction and survival, and that such changes can take place more quickly than was realized.

In other words, the Finch study demonstrated adaptation and the surprisingly short time frames in which it can be observed, even as it occurs naturally without intentional interference. The author of the questions, Jonathan Wells, confused (I suspect intentionally) adaptation with evolution and, by implication, speciation. In the absence of fluency with these concepts the question probably seems to raise a good point and the answer may seem evasive, particularly to somebody with a prior bias against evolution.

I said "unfortunately" earlier because, if we use public opinion of economics and foreign policy as our basis, the average observer apparently has very little capacity or interest in making subtle and nuanced distinctions. Thus I wonder how prepared high school students are to appreciate the answers to these Ten Questions or even, for that matter, if they are capable of understanding what the questions mean.

Then it occurred to me: what a great opportunity these questions are for teaching science! Instead of just giving these answers, or even waiting for a student to come in with them, pre-empt by using the questions and the quest for the answers as the basis for a biology course. In order to address each question the students would need to figure out:

What do the terms used mean?
On what information did the author base the question?
What information would be needed to answer the question?

A good teacher could guide students through this investigative process by asking the right questions and helping to make information sources available, but without explicitly giving away answers. They'll have to read The Beak of the Finch, draw and study genetic family trees, learn about Haeckel and search for references to him, understand and apply the term homology, and so on. All in the context of answering questions about which they may genuinely be curious.

I would argue that students who demonstrate fluency with the concepts and familiarity with the sources but still refuse to budge on religiously based convictions are ultimately far better served than those who are spoon-fed answers; better even perhaps than those who dutifully record data and draw graphs in lab sessions.

Just an idea.

Avatars

2008-08-19T13:10:00.000-07:00

av·a·tar /a-və-ˌtär/ n.

the incarnation of a Hindu deity
an electronic image that represents and is manipulated by a computer user

(Merriam-Webster)

The topic of this post is the 2nd definition, of course. Avatars are something I've been thinking about for a long time in the context of business videoconferencing, and lately also as it applies to education, more specifically to distance learning. I'm bringing it up because it's an illustrative example of how some of the perceived drawbacks of remote teaching can be alleviated or even overcome completely through technology.

The basic idea of an avatar is a computer graphics (CG) image that represents you, in the sense that the actions and words you express through the keyboard and mouse are executed by the avatar. The CG hero you control in a video game is an example of an avatar: you press the "run" key and your avatar runs.

Over the years I've seen vastly more advanced implementations of avatars: ones that respond not just to keyboard input but your actual body and facial movements and the words you say. For example, a camera can track the movement of dozens of points on your face and use that data to draw a face making the same expressions on somebody else's computer. Voice analysis software can augment this process by using the sounds you make to predict the probable shape of your mouth, effectively adding higher resolution to the camera in an area critical to communication.

Furthermore, avatars aren't restricted to cartoon-like characters. With a few good images of your face, your avatar can look like you with nearly photorealistic detail. The computer first uses the movement data to construct a generalized 3D model of facial movement as a wireframe, then stretches/scales that model to match your features, and finally wraps your image over that model to make it look like you. The result is something very much like digital video, but without ever recording or transmitting actual images other than the initial ones (and even that isn't necessary if it has been previously stored.)

Given the near ubiquity of webcams and the success of video, you may wonder why anybody would bother to go through all this. Why not just use actual video? Well, video has some annoying limitations:

Quality video is highly dependent on proper lighting, which is rarely available in typical office or residential environments. In many cases complaints about video conferencing are due not to image resolution but to the shadows that hide facial expressions. Those expressions are a vital part of human (and other species') communication and although we've adapted to the absence of facial expression on the telephone, their tantalizing near-presence in video communications only emphasizes their quasi-absence.
Compression technologies and network capacities continue to improve, but full-motion video still takes up a ghastly amount of bandwidth. Thus the tiny video windows we are accustomed to seeing. Doubling the size of a video window (i.e. doubling each dimension) squares the amount of data required to fill it. And multiple concurrent video streams, i.e. video conference calls, exacerbate the problem super-linearly: the nth participant necessitates 2 * (n-1) additional video streams.
Since camera and screen are two separate devices, real video suffers from the "lack of eye contact" effect: if you are looking at somebody on your screen you are not looking into the camera, which means that at the other end your image appears to be looking off to the side instead of directly at the viewer.
Real-time video is dependent on smooth data feeds: the image starts freezing or skipping if one's internet connect gets laggy or jittery. This in turn is highly disconcerting to the viewer and spoils the illusion of presence.

Avatar technology can alleviate all of these problems:

Lighting (and shading) in a CG image is arbitrary: as long as there is enough light on the source to register movement the reconstructed image can have whatever lighting is desired.
The bandwidth required for an avatar is miniscule compared to that of video. All that needs to be transmitted (or stored) is coordinate data.
The reconstructed avatar does not have to behave exactly like its source: even if you are not looking directly at your camera your avatar can be drawn so that you appear to be making eye contact with your viewers.
CG rendering algorithms can easily "fill in" for missing or late data to maintain the illusion of fluid, natural movement. No jitter or freezing with avatars.

Avatar technology has the ability to not just overcome the limitations of video, but to do some things impossible with video:

With good eye-tracking technology it can be determined where on the screen you are looking. Thus your slight glances in a video conference call can be rendered in your avatar as full headswings, letting others know where you are looking, contributing to the illusion of a live meeting.
It doesn't necessarily have to be you, or the current "you", that is rendered as your avatar. Your avatar can be clean shaven and always wearing a tie, regardless of your current state. Or your avatar could represent a persona you share with others (such as in a customer service environment).
Need a quick break to handle another call or grab a sandwich? Avatars can be put on auto-pilot mode and coast along with natural but unspecific motion. It's sort of like the "mute button" for video conferences (with the same awkward consequences if you take advantage of it too much.)

For all of the above reasons I've long thought that avatars ultimately offer a better solution for video conferencing in a business environment than real video can, but the arguments are equally applicable, or even more so, to distance learning. Certainly the reduced band width requirements make avatars attractive when budgets are constrained. But it's the improved quality of interaction that really makes avatars compelling for education:

Facial expressions, rather than being lost to poor video and shadows, can be exaggerated in effective but cartoon-like ways; an effect that might seem inappropriate in a business meeting.
The teacher's avatar can look at a blackboard or other collaborative workspaces on the screen to point something out, then look back at the student.
A teacher can "look over the shoulders" of multiple students simultaneously without the student being aware of temporary absences resulting from multi-tasking.
Avatars can both provide rich feedback for teachers while giving peace of mind to those who worry about young children and the internet: a child's avatar can be a cartoon representation than conveys emotive expression without exposing the child's actual identity.

Finally, it should be noted that kids today are used to the idea of avatars and will require a much smaller cognitive leap to associate an avatar, even a primitive, unrealistic one, with a live human. We have to be careful not to impose our own adult prejudices and expectations on the intended users of these systems.

I hope designers of distance learning platforms will explore the possibilities for using avatars. It's been a few years since I've looked at this space so I can't recommend specific technologies or vendors, but as I come across things I'll post them.

Teacher vs. Machine

2008-08-15T09:00:00.000-07:00

I'm going to skip the 2nd question from my post about Disrupting Class and address the 3rd question first (or is that second...?): Is a human teacher even a necessary part of the solution?

The short answer, or my short answer anyway, is a qualified "yes". Humans will always be a part of the teaching process, but our definition, or even use, of the word "teacher" will evolve. Unfortunately this question, like gun control and Macs vs. PCs, tends to inspire irrational zealotry on both sides of the debate. So let me phrase it this way: technology has the potential to free up teachers from parts of their jobs they probably find less rewarding, but no matter how far the technology advances...even if by itself it is eventually more effective than teachers of today...it will always be even better when used in conjunction with human teachers. However, I have a couple of provisos:

The role of the "teacher" will likely evolve and branch over time, and we may start using different terms.
For economic reasons we may not always include human teachers in every place where they can improve a technology based interaction; instead we will concentrate them in areas where they can have the greatest impact.

A principle I discovered through my work at General Atlantic is that technologies intended to replace humans usually fail because their designers underestimate or ignore the importance of qualities that are (thus far) uniquely human...qualities such as judgment, discernment, and perception. I'm not saying that machines will never display those qualities on a human level, and I don't want to go down the rathole of debating AI here, but for now the computers don't measure up.

Technologies that do work are those designed not to replace but to augment humans, specifically by offloading tasks requiring precision and repetition. Not only are machines better at these sorts of tasks, but it frees up the humans to focus more exclusively on judgment, discernment, et al. In other words, we're talking about specialization. Often the efficiencies resulting from division of labor mean that fewer humans are needed to complete the same amount of work, but I can't stress enough that it's not because some of the humans have been replaced by machines but that the remaining humans are working more productively.

An example of this distinction is at the checkout counter. Barcode scanners have made checkout clerks vastly more efficient by automating a part of their jobs that required repetition and precision. But attempts to get rid of the human completely and let customers do their own scanning have been largely unsuccessful. Why? Because even with some automation the job required exception handling for things like missing or damaged barcodes, incorrectly labeled items, or even just irrational customers. Most people I've spoken with about their experiences with fully automated checkout report dissatisfaction, and the only place where I've seen continued use of automated checkouts, Home Depot, isn't really 100% automated: they typically have one employee overseeing four checkout stations.

If the supermarket checkout cashier's job requires discernment and judgment, what about the teacher's job? Mark Prensky has suggested that instead of hiring "experts" for our teachers we try to hire "empaths." His point is that understanding one's students, each one individually, is critical to guiding them effectively. Again, I don't want to spend too much time arguing this point. If you are utterly convinced technology can completely replace teachers some day then I'm unlikely to talk you out of it. But the real question is not if technology can replace teachers but if it should. If you think it should, are your reasons based on pedagogy or economics?

However, I do want to rewind a bit to a term I used earlier that is extremely relevant to this topic: specialization. In a broad sense I was referring to the contrasting skill sets of humans and machines, but let's just talk about the humans for a second. To what extent do teachers specialize? By age and subject? Am I missing anything? I suppose we have "special ed" as a separate category, but that's pretty much it.

In other posts I've argued (by implication) that we should differentiate teachers by more than just age, geography, and subject; we should also...or perhaps primarily...be distinguishing teachers by the specific learning styles they are most suited to address. But I also want to point out that teachers essentially never specialize by function. That is, every teacher of every subject and every grade has basically the same duties: conducting class, creating handouts, giving tests, scoring tests, averaging grades, delivering lectures, showing videos, helping stragglers, writing lesson plans, etc. etc. etc.

I'm going to go out on a limb here and state that it's insane we don't ask our teachers to specialize any more than we do. Imagine talking to an autoworker and finding out that he specialized in "red sedans". Or a doctor who specialized in "People born in even numbered years." If you then asked which particular part of building cars or administering medical care they performed, you'd be surprised and disheartened to hear a puzzled "Well...all of it."

There are really two reasons to encourage specialization in any profession:

Efficiencies. David Ricardo's Law of Comparative Advantage is usually used in the context of international trade, but it explains the value of specialization in general and the value gained by it even when one party is more proficient at all tasks.
Satisfaction. Some tasks are essentially more interesting and rewarding than others. In most professions one benefit of seniority (whether based on time or merit) is gradual relief from the more onerous tasks.

Typically there is a high correlation between the appeal of a task and its value-add, so these two factors are not independent: senior workers get to do the interesting and valuable work; junior employees do the easier drudge work. We see this to some extent in post-secondary education, but not at all in K-12 education. I'm going to address this point in more detail in the last post of this series, but for now to circle back around and tie technology back into the discussion let me both reiterate and unify my main points:

The purpose of specialization is to improve overall system efficiency.
The most effective use of technology is in the context of specialization: computers and humans each focus on the tasks for which they are best suited.
Even really primitive technology can, if properly applied, make humans more productive.
Conversely, no matter how good the technology is, participation by humans can make it more effective.
For economic reasons our goal for deciding the right allocation of our resources, distinguishing not only between human and machines but also between humans with different strengths, should not be to provide the optimal solution in each case, but to devise an overall plan that provides the greatest cumulative benefit. Anything else inherently ignores the fact that resources are always limited, regardless of how high or low that limit is.

Fantastic Contraptions

2008-08-12T09:29:00.000-07:00

Jeremy Bloom recently alerted me to Fantastic Contraptions, a remarkably creative web site to which I've become addicted of late. It's a kind of game where you assemble machines to solve puzzles. I won't try to describe it beyond that...you should go and poke around at it. But here are the things I love about it:

Incremental progress: the outcome of your attempts to solve problems do not have the binary outcome of just "succeed" or "fail". Rather, there are degrees of failure (or, alternately, degrees of near success). This has the effect of breaking a problem down into many small steps, and the completion of each feels like progress.
Iterative assessment: it's not only simple to make frequent, repeated attempts at solving the puzzles, the system is designed to encourage that approach. Thus you can make tiny changes and try again as many times as you want.
Rich feedback: when your attempts fail you can see exactly what went wrong and what didn't work. The iterative nature makes this aspect even more powerful: you can repeatedly re-play things until you understand the problem.
Multiple solutions: there is no one "correct" solution to each puzzle. All that matters is successful completion; but there are often radically different ways to approach each one.
User extensible: you can create your own puzzles (levels) and share them with other people.
Social interaction: the associated forum provides a community of enthusiasts. As much fun as it is to solve the puzzles, it's also interesting to see how other people solve them, sometimes in truly astonishing ways.

What I've been thinking about since discovering Fantastic Contraptions is how it could be used in an intentional way with kids as a learning tool. In particular I'm thinking about the studies in which students recite Newton's laws yet show an inability to apply them to simple phenomena accurately. Could the use of this game help kids more intuitively learn Newtonian physics, which might in turn help them better internalize and contextualize the formal laws?

I may have a chance to test this out this fall with some 5th graders. I'll report back with what I discover.

Kunskapsskolan

2008-08-12T08:08:00.000-07:00

I was talking with Graham Glass the other day about his Edu 2.0 venture, and he told me about some innovative schools in Sweden that have arisen out of their voucher-driven educational system. (Sweden has been aggressively supportive of vouchers since the early 90's.)

Some poking around on the web unearthed the name of these schools, "Kunskapsskolan", which translates roughly as "Knowledge Schools". Here are some pithy quotes from wikipedia:

"It puts much more responsibility in the hands of the pupil by removing classrooms and letting the pupil choose when, where, and what to participate in."
"...the pupil works on his own level, which is measured in "steps" ranging from 1-35.
The pupil will work by themselves with support from the teachers for their own education plan."
"Every morning the pupils have to take up their logbook and plan the day. The pupil themselves plan the day like a to do list."
"Kunskapsskolan don't put their students in classes with a teacher lecturing the pupils."
"The education at Kunskapsskolan is very IT based, instead of giving the pupils a couple of books to work with, they have to find the steps on a website called kunskapsporten and print out the step they are working with."

Consider this to be foreshadowing of some upcoming posts.

The Classroom Paradox

2008-08-11T11:29:00.000-07:00

In a previous post I tried to explain one powerful argument for distance learning (i.e. that it allows for more optional allocation of teachers and students) and ended with the promise to answer three questions. Let's take a look at the first question: Can technology make distance learning and online learning effective enough that the advantages of location-blind clustering outweigh the disadvantages of not having a live, face-to-face teacher?

The hard part about answering this question is that distance learning, both the technology and our experience with it, is still so nascent that any discussion about its ability to replicate live interaction depends to some degree on faith. I think Clay Christensen would argue that its the nature of disruptive technologies that many people cannot see their promise when they are still nascent. When the first transistor radios came out it wasn't obvious that solid state audio components would eventually sound as good as vacuum tube radios, let alone live performances. It wasn't obvious to everybody from the first personal computers that they would eventually be as powerful as they are. And it may not be obvious looking at the current state-of-the-art in distance learning that it's going to keep improving exponentially....while traditional classrooms are not.

So instead of trying to convince you that distance learning will close the gap on face-to-face instruction in terms of "quality of personal interaction" lets talk about how big that gap is. Or, more accurately, how big that gap can be just to break even with the logistical handicap of the classroom model.

In the traditional classroom model we group kids by age and ignore, for the most part, aptitude and learning style and stage. Let's give this model the benefit of the doubt and assume, or at least pretend, that kids in the same class start off the year at the same stage, meaning that in a given subject they're all at the same point with the same material mastered.

In the graphic above, imagine that each colored bar represents a particular student and their progress through a given topic over a fixed period of time. For example, this could represent algebra over the course of a semester. The plan is for the students to progress together all the way across the chart by the end of the semester.

As we all know, however, kids do learn at different rates, and this graphic represents that by having differing lengths for the bars. The blue bar at the top represents the "smartest" and/or hardest working kid, whereas another blue bar a couple rows down represents a "slower" kid.

The teacher, of course, has to teach all of the kids and therefore must strike a balance in addressing the whole class. That vertical red bar represents the average position of the entire class. That is the approximate target stage for the teacher.

So far that seems ok. Sure, the bars aren't all the same length and at the end of the semester some kids will be slightly ahead, but for the most part this works, right?

Not quite. The problem is that the rate of learning for each student varies with their distance from that red line. That smart kid is so far ahead of the level at which the class is taught that she isn't learning anything new, or at least not as much as she would if the class were geared to her intelligence. Meanwhile the slow kid also isn't learning as much; the material is just a bit too advanced and while he may grasp some of it he's also confused and frustrated because he's just a little bit behind his peers.

So the smart kid is still learning, and learning faster than the average kids, but not anywhere near her capacity. And the slow kid is still learning, but slower than the average kids, and also below his capacity. So what happens to our graphic? The bars continue to grow, but progress gets slower and slower and the gaps get wider and wider. Even the "average" kids end up suffering from this effect because overall classroom progress is slower than its theoretical maximum, so the average point of instruction (the red line) is constantly to the left of where it could be, and falls further and further behind.

You don't even have to believe in "learning style" to understand this principle. All you have to do is acknowledge that some kids learn faster than others, for whatever reasons, and that teachers can't possibly customize the classroom for every kid in the class, to realize that this effect will inevitably occur.

You might argue that this model predicts a breakdown or stagnation in learning that we don't actually observe: countless classes do in fact make it through the end of the semester with most or all of the kids apparently having learned the material. So where's the failure? The failure comes from the fact that we have already accounted for this process by setting relatively low expectations in the first place. We use the success or failures of the past to set objectives for the present, and the inefficiencies of the classroom/factory model have determined the rate of learning that we deem acceptable.

The only way to counter this effect is to keep classrooms "tightly clustered" in terms of stage, and the only way to do that is to have a very large pool of students. And of course this is why some schools "track" students roughly by ability. That's not a very granular method, but certainly it's more effective than random distributions.

But note that I said tightly clustered in terms of stage and not by ability. It's not important that kids have the same aptitude and learn at the same rate, as long as that at any given moment the kids in a class are at the same stage in the subject being taught. Now, you may be thinking, "Sure, but in order to keep them at the same stage they have to keep learning at the same rate." But that's because you're still thinking about physical classrooms and the logistical requirement of creating classes for entire semesters or years. If that's the case then, yeah, you want to try to keep those bars close by clustering kids by ability. But what if you can reallocate students on a weekly, daily, or hourly basis? What if you can find a class and a teacher for each kid at any moment that exactly meets his or her needs? How fast are those bars going to move across the graph?

And that, of course, is the real promise and power of distance learning. By working with a vast pool of students...many orders of magnitude larger than even the largest physical school...and completely dispensing with geographic constraints, you make it possible, at least in theory, to have highly, highly optimized allocations.

So that's why distance learning is potentially much better than physical classroom learning. Now let's get back to the question about quality-of-interaction and look at why it might be worse. The main criticism tends to be one of distance learning's intrinsic properties: the distance. The argument is: "The personal connection is vital for good teaching, and you can't have the same quality of personal connection over the Internet."

While I won't say this argument is fallacious, I do believe: a) it's not as true as some people think, b) as we stated at the beginning, it's becoming less true all the time, and c) the disadvantages are still vastly preferable to the disadvantages of the Classroom Paradox described above. Three things to keep in mind:

The generation(s) we're talking about, the one in school today and the future ones not yet in school, are not like you and me: they are used to having active social lives via telecommunications. They find nothing surprising about being "good friends" with somebody they've never actually met. They are the "Digital Natives" that Marc Prensky describes in his book, Don't Bother Me Mom, I'm Learning!.
Technologies for remote communication and collaboration are improving and becoming more ubiquitous constantly. Video conferencing, virtual blackboards, voice-over-IP, collaborative editing, avatars, and other interesting technologies are getting better all the time, and people everywhere, particularly young people, are finding their use to be completely normal.
Lastly, let's not over-romanticize physical classrooms. Let's not forget that two dozen kids in a confined space comes with its own set of problems and distractions. As one observer pointed out (sorry, forgot where I saw this reference; I'll attribute it when I figure it out): "There are no discipline problems in a virtual classroom."

Given all these arguments, my personal conclusion is that live classroom teaching does currently have some net advantage in terms of quality of personal interaction. But that advantage will steadily erode and eventually disappear completely, while the net disadvantage of live classroom teaching, i.e. the physical proximity requirement, already vastly outweighs the advantages. And the scale will just continue to tip in favor of distance learning.

Seeking feedback from teachers

2008-08-11T11:07:00.000-07:00

I've lately become enamored with the idea that a free, on-line tool for creating classroom materials (handouts, worksheets, etc.) would be a Good Thing. I've started talking up the idea in various circles, and have hopes (delusions?) that I'll eventually convince somebody to build this...even if that somebody I convince turns out to be myself.

My ideas about what this tool would look like are inspired mostly by my own short stint as a teacher, and by notions of how I _assume_ other teachers work, but I would like to hear from actual teachers about what their jobs are like. This will help me understand better what features would be useful, and in turn make my argument more convincing as I evangelize the idea to various organizations.

If you would like to help, please take a few minutes to answer the questions below and either post them to this blog or email them to david.rea@gmail.com. Please be as elaborate and descriptive as you wish in your answers; I am looking to really understand how other teachers work. And if you would like to be updated on this project as it progresses just say so (and include the email address you would like me to use). Thanks!

1) How much time per week do you spend authoring and editing materials that you hand out to students? And what tool(s) do you use? (I.e. typewriter, Microsoft Word, Photoshop, etc.)

2) To what extent do you re-use materials from previous semesters/years? Do you update them or just use the exact same ones? What factors affect your decision to recycle/edit/recreate?

3) Do you ever get materials from other teachers/sources, either from the Internet or colleagues you know? What factors make this work better or worse?

4) What's the worst/slowest/hardest part of creating materials? (Examples might be: "Formatting equations correctly", "Copying vocabulary definitions", "Just thinking up good questions", etc. Whichever part you dread.)

5) What ages/subjects/demographics do you teach?

6) Do you know of any existing tools designed specifically to help teachers create worksheets and handouts?

Again, thanks! And please forward this to your colleagues!

Disrupting Class

2008-08-06T13:37:00.000-07:00

I recently read Clay Christensen's Disrupting Class and it's now on my must-read list for anybody interested in educational reform. Christensen is, of course, the author of The Innovator's Dilemma, the acclaimed book about why large organizations are unable to adapt to disruptive technology changes. This new book, co-authored with Michael Horn and Curtis Johnson, applies the same models to the "business" of education. There are many takeaways in this book, but the ideas I found most compelling relate to the argument for distance learning. I'll try to explain that argument over several posts, but the first step is to understand how much the standard educational model is designed not to provide the most effective educational model but to solve a simple logistical problem, unrelated to learning.

Given X number of students, and Y number of teachers (assuming that X is much bigger than Y) the first problem an administrator must solve is how, at any given time, to divide up X into Y number of groups. There are many variables (i.e. dimensions) by which you could theoretically cluster students, such as age, subject, interest, stage, learning style, skin color, aptitude, gender, location, alphabetical order, language, etc. The order of precedence of these dimensions will vary depending on what outcome you are optimizing for. For example, if you are a local school district optimizing for efficient allocation of teachers and classrooms you will cluster your students by age and subject. If you try to cluster students by aptitude, stage, and learning style you will typically have too many groups, of wildly varied size, to allocate your teaching resources efficiently.

With larger pools of students you can utilize more dimensions, and/or with more granularity. While a small rural school might reduce granularity and combine grades, a larger school district might "track" students by ability into two different groups. A very large urban school system might further sub-divide into 3 aptitude tracks and even have separate classes for English and Spanish speakers. But the point is this: the way we divide up our student body is determined more by budgetary and logistical considerations than by sounds pedagogy.

However, if you can get rid of the requirement of geographic proximity the potential student pool becomes so large that you could, in theory, utilize only the dimensions that are actually relevant to learning (such as aptitude, interest, learning style, and stage) and completely dispense with age and location, which are utterly irrelevant to learning, as factors.

The exact same logic, by the way, applies to allocation of teachers. Some teachers are brilliant lecturers or storytellers; others are excellent tutors. Some teachers work great with visual learners; others work well with mathematical-logical thinkers (to borrow terms from Gardner). But in our geography-limited system of schools we simply cannot factor in these differences; for practical reasons we can at best allocate teachers by age and subject.

Thus it should be obvious that by ignoring physical proximity we could have much more effective and efficient allocations of students and teachers. This line of reasoning points toward distance learning (with a teacher) and online learning (without a teacher) as possible solutions, which in turn presents three important questions:

Can technology make distance learning and online learning effective enough that the advantages of location-blind clustering outweigh the disadvantages of not having a live, face-to-face teacher? (Addressed here.)
Even if the technology is capable, is there any pragmatic hope for technology to overcome the political and cultural obstacles to such a dramatic change?
Is a human teacher even a necessary part of the solution? (Addressed here.)

I'll address each of those 3 questions in future posts.