So did y’all know that Jing has a monthly bandwidth limit of 2GB? Yeah, neither did I. So it turns out that the Jing walkthrough I’ve been passing around maxed out a bit ago.

So I went through the labor intensive process of using Slideshare to simulate a screencast of Waterfeed. The upside is there is no bandwidth cap. So please pass this link around instead of the Jing one. Jing is dead to me.

(and if you want that link in plain text: http://www.slideshare.net/MichaelCaulfield/waterfeed-water106-presentation)

For those of you who have no idea what I’m talking about, welcome to the Waterfeed party!  Start by watching the short video below. It details a really cool project we’re putting together and we want your participation. It’s A-Game stuff, and you won’t be disappointed.

Vine is one of the addictions my daughter and I currently share. Somewhat predictably, the six second format and lack of  an editing function acts as generative constraints that brings out the creativity in people, and it’s just fascinating to see what people come up with.

If you’re looking for an example of how to weave this into your class, take a look at GE’s “Six Second Science” Vines, where people were asked to use the format to demonstrate science. (They do get a bit old after a while, but you can honestly watch more of these than you think in a row).

I’m thinking through how we might use this sort of thing in Water106. I like the idea of a massive “contest” which is really just an excuse to muck around. Perhaps something open-ended like “Do a Vine that demonstrates something you learned about water in this class?” Or maybe just “Water. Do a Vine about it?”

Thoughts?

So I admit that I was initially confused by what the President could do *now* about his education plan. With the current Congress, nothing is getting passed. There are some elements in the plan which can be done through pure executive power, but most of it requires legislation. So why now? How does this announcement matter today?

The answer is in the first section of the fact sheet:

Before the 2015 school year, the Department of Education will develop a new ratings system to help students compare the value offered by colleges and encourage colleges to improve…Over the next four years, the Department of Education will refine these measurements, while colleges have an opportunity to improve their performance and ratings. The Administration will seek legislation using this new rating system to transform the way federal aid is awarded to colleges once the ratings are well developed. Students attending high-performing colleges could receive larger Pell Grants and more affordable student loans.

I don’t know if you caught that, but here’s what’s going on. The administration has the executive authority to collect these numbers on effectiveness and define the formula. What it is saying is that over the next four years it’s going to be tracking schools and ranking them. The plan is to propose them as a funding formula in the near term, and steer Title IV funds towards best “bang for buck” schools.

So let’s say the President doesn’t get this into the next Higher Education Act reauthorization (he won’t). When he fails, the numbers don’t go away. They keep getting compiled and refined, waiting…

When 2017 rolls around, what happens? Well, it’s potentially a whole new Congress. It’s potentially a Senate without a filibuster. And this will be proposed again, based on these numbers.

So here’s the deal — as these rankings are developed, and you find your institution is near the bottom (or even the middle) you have a choice. You can assume that the political reality after the next presidential election will look like it does today, and slough off these numbers being compiled about you as insignificant. Or you can take the view that there’s enough of a chance that environment may be favorable to the bill’s passage in a couple years that you’d better take the numbers seriously.

I actually think this is a good thing, but perhaps I’ll leave that to a later post. The question most administrators and state legislators have to ask themselves today is “Do I feel lucky?” I’m not sure how they will answer that.

I love working on ideas and projects that no one else seems to be doing, but the best moments in my professional career have been when I’ve discovered that what I’m doing is not so original at all. Fresh off of an interview with Jim Groom, I’m reminded of that fateful moment in 2007 where I first bumped into Stephen Downes’ OLDaily newsletter and realized that the thinking I’d been doing on what an “Inverted LMS” looked like had already been thought through as a “Personal Learning Environment”. I also remember the moment when I started New Hampshire’s first blog dedicated exclusively to covering Congressional District Two’s race from a progressive standpoint — only to discover that in the same month two other people had had the same idea.

In both those cases the fact that many people were working on similar ideas separately was a sign that we all were plugged into something much greater. And so it’s with great excitement I stumble today on the FemTechNet DOCC running this fall, which looks a lot like ds106, independently discovered.

I’m going to dig into documents on the DOCC to see what they might have that could inform the design of Water106. I’m guessing the people running it already know of ds106, the original MOOCs, and UBC’s Latin American Literature Wikipedia Project, but in case they are as unaware of those things as I was of the DOCC, I’ve added links here.

(And incidentally, I’m not so much a fan of the acronym DOCC, but it’s a better name for this structure than MANIC)

I’m working on Waterfeed, one of the activities of the Spring 2014 experiment we’ll be running, where we run a cMOOC/ds106 experience over multiple classes as an integrative layer. And something happened this morning when I was looking at the feed that seemed like a good opportunity to talk about how errors in a network get corrected. It’s also a chance for me to go on a mini-rant about the “We want a culture of Producers, not Consumers!” silliness that infects EdTech. Do you want “a culture of Writers, not Readers!”? of “Musicians, not Dancers!”? of “Typewriter Makers, not Typists!”?

No, of course not. At the very least, you want a culture of *better* producers, and *better* consumers. And frankly, I’m not sure you want to divide the world into those categories at all. It’s kind of elitist, capitalist bullcrap based on the idea that producers “give” and consumers “take”. That producers bring new stuff into the world which consumers deplete. Nothing could be further from the truth, which is why I prefer to use the terms reading and authoring when talking about digital and literary products, as these terms emphasize the important, active role of the reader in the process of creation and dissemination of art and science.

So, in any case, this five minute screencast shows what good digital reading looks like, why we need it, and why “production” without reading is a bad, bad thing.

People who read this blog know I’ve been talking about Water106 for a while and about the possibility of applying a ds106 model to an issues-based course for even longer.

For those that haven’t been following the idea as it has been developed, however, I’ve written up a whitepaper, suitable for sharing with faculty or administration that you might want to rope into the project. It covers the story up to now, and I think renders the reasoning behind it more comprehensible.

I will say that the more I work on this, the more excited I get. Part of that excitement is that we are not only educating students, but modeling a way of working together cooperatively that students can bring into the workplaces they head off to (or the ones they currently work at). This is, of course, true of ds106 and cMOOCs as well, I’ve just never designed a networked learning project of this magnitude. Instead of preparing students for business, we’re preparing students to change business.

Anyway, take the time to read the whitepaper linked below if you’re interested. It looks like a lot of pages, but it’s really mostly screenshots.

Having students summarize readings for someone else is one of the great teaching techniques. We see it in Peer Instruction, we see it in the effects of team-based learning, we see it in the beneficial effects of tutoring on a tutor’s understanding.

At the same time, such activities are often inauthentic. WaterFeed is a part of the Water106 project I am working on that tries to remake student summary into a meaningful endeavor. The way WaterFeed works is the backend of the site pulls stories on water policy, technology, and science from more than 80 feeds daily. These are stored as drafts. Students are then encouraged to go in and find stories in the drafts they would like to summarize. Those summaries, when done, are pushed out to the WaterFeed blog where they can be used by both professionals and students to stay up on the latest news and research findings. When these summaries are published, they move out of drafts, ensuring that students are not all covering the same few stories.

I can’t embed a Jing Screencast on this blog, but a five minute presentation of how it works is here. I’m particularly excited because it seems to me that this is the sort of activity that could be directly applied to any number of courses from many different disciplines.

Yesterday I put together a spreadsheet and looked at two different approaches to income-based repayment — a current federal program (Pay-as-You-Earn) and Oregon’s proposed Pay it Forward. It did not end well for Pay-it-Forward, although I’m still waiting for someone more familiar with modeling these policies to pick up the torch and check/fix my models.

Now that I have the spreadsheet coded, it’s easy to look at the effect of assumptions. Yesterday we saw what a huge difference getting to degree in five years makes compared to degree in four, and how that change in assumptions generally makes Pay it Forward one of the more expensive options. Today I want to show the effect of dependents. Here is a person who earns an average of $30,000 for the rest of their life. (Again, you want more complex scenarios, then read the original article and download the spreadsheet.)

You’ll notice $30,000 a year is the break-even point for single people with no children. Below this Pay-as-You-Earn, with its progressive structure, handily outperforms Pay it Forward. Above $30k/yr, Pay it Forward starts to provide a low lifetime cost to the loan.

But that’s for single people with no children. Add a child into the equation, and the story changes pretty dramatically:

Because Pay-as-You-Earn is tied to discretionary income, and discretionary income is impacted by children, Pay-as-You Earn ends up a better deal here for single parents (I’m not modelling married parents because it’s a hassle to code the spreadsheet).

How much does it impact it? So much that Pay-it-Forward becomes three times as expensive as Pay-as-You-Earn. Three times.

So what ends up being the pivot point for the single parent? At what level of income does Pay it Forward become a better deal than Pay-as-You-Earn? It’s actually $47k/yr.

This means that people making a lifetime average of $47k/yr or less are better off going with the federal program if they are single parents. Interestingly, this is also the break-even point with lifetime cost of a standard loan (although again, the high payments on a 10 year plan are an issue here).

Again, I want to stress this is a rough pass at these figures. Tweet these around if you must, but the point here is not to “win” the argument, but to convince people to start doing the math.

(With apologies to Jerry Garcia).

OK, so here’s the thing. I just want someone to model what the Pay it Forward, Pay it Back plan looks like in a variety of cases compared to actual existing alternatives. That would seem an easy thing, something that legislatures and policy houses would do before taking positions on it. You’d think, at least.

I come off as a person who is against Pay It Forward. I’m not. I’m against the Pay It Forward tulip madness that has engulfed otherwise reasonable people. There are some basic questions you should be able to answer before talking policy positions. Nobody I can find can answer those questions. That should freak you the heck out.

One of those questions is, “Hey, how do the total payments under PiF compare to payments under other government programs like Pay as You Earn, the existing federal program that caps payment at 10% of discretionary income, or the Public Service Loan Forgiveness Program which forgives loan balances after 10 years in exchange for public service?”

Since no one seems to be interested in answering this, I went and dug up the formulas and wrote a spreadsheet. I modeled two scenarios — a single parent public school teacher, and a single engineer. I chose single people because the math for dual incomes gets too involved.

Here’s the spreadsheet. If you find something I’ve done wrong in it, I’m not surprised (read the post title again). But if you do find an error, please do me a favor and rather than carping about it in the comments, fix the spreadsheet and show how it alters the model.

Public School Teacher, one kid.

Ok, so the first thing I modeled was a public school teacher with one kid and no spouse. I compared what payments looked like under the following scenarios:

• Standard 10 year loan on $34,000 of debt.
• Standard 25 year loan on $34,000 of debt
• Pay it Forward, Pay it Back
• Pay-as-you-earn Federal program
• Pay as you-earn Federal program with Public Service forgiveness

I used the following teacher pay model, pulled from NCES data (and simplified to 5 year increments):

Then I modeled two scenarios. a degree finished in four years, and a degree finished in five. You should note that for a person to get an education degree in four years is pretty rare, but I was interested in how changing this assumption affected things. So first, here’s our time-to-degree superstar, coming in at four years:

So this is payment over the life of a loan, 4.5% interest (higher than currently, but buffering a bit) on $34,000, which I’ve heard kicked around as median student debt held by borrowers (This sounds slightly off to me, but benefit of the doubt. I’m too tired to track down references).

What we can see in our four-year scenario is what a difference that public service loan forgiveness policy makes. Pay it Forward is 75% more expensive than PYE with forgiveness.  In this scenario, Pay it Forward essentially takes a $15,000 subsidy for public school teachers and throws it out the window. I haven’t modeled other government workers, police, military personnel, federal research scientists, community college professors, and state engineers, but I’d assume there are some lost subsidies there too.

What’s more striking, however, is how even in this scenario the Pay it Forward option is not significantly off from the ten year loan or the pay as you earn options without the Public Service benefit.

That story gets more pronounced when the student takes a 5th year to get to degree, accruing another 0.75% tax on their income. Here’s the 5-yr scenario:

Now Pay it Forward is about the same as Pay as You Earn without the Public Service forgiveness. With the forgiveness Pay it Forward significantly more than double the lowest option. Most interestingly , Pay it Forward is more than the standard 10 year loan (although of course the 10 year loan has higher monthly payments than the 24-year PiF program, which is a factor).

It’s surprisingly hard to see a clear benefit here. I could jack up the loan amount, which would up the standard loans, but even then the totals of the two Pay-as-You-Earn options would be the same, since they are based on income, not loan amount. For your median public school teacher, Pay it Forward appears to be a really lousy deal.

The Engineer Model

What about for mechanical engineers? Here I got lazy. I found median starting income online and then increased it over the twenty-five year window assuming promotions to senior engineer and then some form of technical management. It’s clearly not as solid as the teaching numbers, but again, look at the post title. We want to model high flyers here. (Also, I live in Camas, Washington, which is engineer central because of the chip plant, and I will tell you these numbers are credible).

Here, if the engineer was eligible for Pay-as-You-Earn, it would cost something like $142,0000. Needless to say, Pay-as-You-Earn is off the table. Because Pay-as-You-Earn is off the table, public service forgiveness doesn’t mean a lot either — you’re going to finish your loans in 10 years anyway, so I ten year forgiveness policy is superfluous. OK, what does this look like? Here’s the 4-year model:

Pay it Forward here is a bit over 50% more over life of loan than the standard 10 year option. The difference between PiF and the 25 year loan is actually less than I thought it would be.  Of course, again, this is impacted by a five year time to degree:

This was the big surprise to me — how in the more realistic five-year model PiF becomes RISD–level expensive. Take your breath away expensive. With these assumptions, it’s easy to see why an engineer might opt for a school with a more standard loan program, and that makes me worry about our state research schools facing an engineer exodus.

So on this model, again, no.

The Strengths of Pay it Forward

Assuming my math is correct and this is a raw deal for both the struggling public schoolteacher and the high-flying engineer, why is it so compelling? (And I do agree it’s compelling, even after looking at a total cost that was in both cases DOUBLE the best currently available option).

There are a couple reasons. Lack of familiarity with current options is one. People just don’t know about the newer options for repayment. When testing a drug you don’t model against nothing, you model against the best possible available treatment. Same is true in public policy, but where anyone has bothered to model anything at all they’ve modelled against a student choosing the worst possible options for repayment. (And to be honest, most people have not even modelled that).

But I *know* these options, and I still feel the pull of it. Why?

• Simplicity. This comes through in the spreadsheet creation. Creating the current options was annoying, and nearly impossible for joint filers. Pay it Forward is math you can do in your head.
• Highlights that college pays off. No matter what scare story Forbes is running this week, you are still better off going to college and accruing debt than not going to college. But scare story after scare story, sometimes from well-meaning people, has scared debt-averse populations away from college (which, guess what, are generally also minority and working class populations). In this paradigm, you might pay more (a *lot* more), but it’s not “debt”, so it’s OK. That opens college up to people overly worried about debt.
• There from the beginning. Here’s a big one — these repayment plans the government has? You don’t even get to decide to opt into them until your out of school. On top of that, they are portrayed as “hardship” plans. So when you are considering school, we can’t even talk about these. We have to present you the vanilla plan that is your worst possible plan. Add to that the fact that we are advertising a sticker price significantly higher than what you’ll actually pay, and we are doing everything in our power to put the worst possible spin on your loans.

In other words, the genius of Pay it Forward is in its rhetoric, not its math.

Now assuming the work I did above is right (which, let’s face it, is a big assumption) — is that rhetoric worth paying *double* what you might pay otherwise? Is it worth paying an extra $25,000 to $40,000 for school? I’d say no.

But the rhetoric and the simplicity is worth a look. A lot more people should be modelling this against current options. But, in truth, the current options should be much easier to model.

I’ve just finished reading Susan Cain’s Quiet, which is a must-read for anyone in instructional design. (And by must-read I don’t mean you should read it because you will like it, but that you should read it because to not read it would be negligent: this book will open your eyes to an educational system increasingly demanding extroversion for success).

There’s a lot of thoughts swirling around my head right now, but one of the more interesting ones is this: why have we wasted so much time talking about learning styles (which don’t seem to matter much in practice with success at classroom activity) and made so little time for talking about the introversion/extroversion spectrum of personality, which seems, in my experience at least, to matter quite a lot?

I’m going to guess it’s because there’s money in coming up with Netflixy learning styles solutions, but there’s not much in deconstructing the assumptions inherent in group work. Other ideas?

UPDATE:

Giulia F. takes me to task for simplifying the learning styles issue. I agree! I’ve written about this issue with more subtlety before, and was rushing to get this post done before my morning commute. In any case here is my response to Giulia:

Thanks Giulia — I often speak in a shorthand about learning styles that doesn’t always capture the complexity of the issue (witness the amazingly convoluted sentence at about them in this post). I believe, in fact, that my position on the subject is somewhat closer to Kolb’s, who I seem to remember saying at one point that of course reading literature was going to require one set of preference-independent skills and doing math another. And that’s largely the rub — the impact of those authentic barriers tends to outweigh the impact of our arbitrary ones.

That doesn’t mean that we should not address the arbitrary barriers, but that the way in which this has been presented and implemented has been just this side of astrology in many cases. The focus on “styles” trivializes deeper issues that students are having engaging with the course. We can’t deal with these issues without a fundamental rethink about what education is about.

So I agree that what we really need when we look at both accessibility and these issues of introversion and “styles” is a universal design approach. And the issue becomes how we accommodate multiple routes to participation while both incorporating smart design based on research and while preventing the complete fracturing of the educational community we are attempting to build around a common experience.

Online approaches, from the earliest Usenet groups to the latest cMOOC or ds106 experience, have some lessons for us there. And, admittedly, display some blind spots as well. I didn’t post my massive post on what universal access looks like on the intro/extro-version spectrum, but the upshot is that if you imagine a workplace that values the work of both introverts and extroverts that you can work back pretty directly to a model for teaching. I hope to cover that in a future, meatier post.