Some Thoughts on OER

Last week I had the good fortune to co-keynote the Northeast Regional OER Summit at UMass Amherst. My counterpart keynoter was Don Kilburn, the current CEO of UMassOnline and former CEO of Pearson North America. We each gave brief talks, followed by a conversation facilitated by UMass Amherst’s Marilyn Billings. It was a lively discussion that inspired a lot of passionate debate on Twitter. That, in turn, inspired requests for more information about the conversation from people who weren’t able to come. So this is my recap.

Don’s talk

Don was there in his UMassOnline capacity but he spoke from his perspective as a long-time senior executive at a major textbook publisher. From the beginning, it was clear that having Don on stage would be both potentially interesting and inevitably fraught. Many folks in the OER community have a visceral negative reaction to the way of thinking and the kind of language that Don employs instinctively due to his particular professional history. And Don, for his part, didn’t seem to have had a whole lot of exposure to or understanding of the audience he would be addressing. The most cringe-worthy moment was when he trotted out the old “free as in puppy” chestnut as if it were a novel statement and not something that the OER community, and the open source community before it, had heard ad nauseam for at least a decade. That cultural clash between the audience and the opening speaker…resulted in the kinds of tweets that you would expect, and appeared to have an outsized influence on the way a vocal segment of the audience reacted to the whole conversation.

That’s unfortunate, in part because Don knows a lot that could be useful to people who want to learn how to be more effective at driving OER adoption and understanding how a broad cross-section of faculty approach curricular materials adoption in general. He knows about the many experiments, both successful and unsuccessful, that the textbook industry has tried in order to figure out which value propositions persuade faculty to adopt curricular materials. He knows what’s happening in the market right now, how the publishers think, where they are gaining traction, and where they are struggling. Whether or not you agree with him, he can provide useful intel that is normally inaccessible to academics.

I won’t summarize his talk here, but since the first part of my talk built off of Don’s, you’ll hopefully get a rough sense of the ground he covered through my summary of my own talk.

My talk, part one

I’m not going to recap the discussion in strict chronological order. Instead, I’ll address the piece of my talk that built off of Don’s now and circle back to the other part—which was really the main part of my talk—later. In retrospect, the conversation after the talk provides some good context for understanding the main point I was trying to make.

A lot of Don’s talk was about how curricular materials prices are coming down and how the industry is trying to establish the value of its product in the face of this change. This seemed like a good place for me to pick up, since the most common argument for OER is about affordability There is absolutely no question that the value of base informational content—the part of a textbook that could easily be replaced by a Wikipedia article, for example—has commoditized. This is one reason why textbook prices are coming down. (I could tell another story about used books and rentals and Amazon and Chegg, but the two narratives are really just two sides of the same coin.)

In the curricular materials markets, there are two pricing bands that are beginning to emerge. The first one is in the $10 to $40 range, and it is often presented as either a cheap version of the print textbook—a black and white softcover, for example—or something close to a direct digital replacement of the book. The other band, in the $60 to $100 range, tends to have products with lots of formative assessments, student and instructor dashboards, nudges and reminders, and maybe adaptive capabilities. Here, publishers are trying to establish a different value proposition from the print textbook. The “courseware” products that typically inhabit this price band can provide both students and instructors with a lot more information about how the students are doing, whether they are coming prepared to class, and where they need help. I have written several posts about these two competing value propositions, labeling them as “good enough” versus “better enough”.

As long as those two value propositions dominate the way in which curricular material choices are framed for (and by) the faculty, they will also frame the way that OER are valued. And I mean that partly in economic terms, since not all OER are cost-free to the students and none are cost-free to the creators and maintainers. If the Wikipedia-like portions of the textbook have little to no economic value, then what else are students paying for and how much should they have to pay for it? How much is professional curation—in the form of scope and sequence—worth? How much is it worth to have somebody align learning objectives, assessment questions, and the informational content? To keep the content up-to-date? To provide frequent, auto-graded or easy-to-grade formative assessments? To provide dashboards that show progress on those assessments? To provide adaptive learning tools as differentiated instruction aids? There is no one correct answer for each of these questions, but now at least the pricing is starting to become transparent enough and product options unbundled enough that it is possible to answer them. The affordability problem, while not yet solved, is moving in the right direction. Because unbundling is part of this movement, educators can start making more fine-grained, student-centered choices about any potential trade-offs between accessibility and effectiveness.

In my view, the OER community needs to become more sophisticated in its discussions of these trade-offs and more respectful of the individual decisions faculty make as they try to find the right trade-offs for their particular contexts. “Free as in puppy” may be glib, but that doesn’t mean it’s completely false. On the one hand, I don’t know anybody who got a free puppy that didn’t…you know…already want a puppy and know that puppies require care and feeding. On the other hand, some people do underestimate the amount of care and feeding a puppy requires. If you want to make sure that those puppies don’t get given to a shelter or abandoned at the side of some road, you need to make sure that somebody is prepared to be responsible for them. You need a puppy sustainability strategy. Some puppies are more work than others, and some families are more prepared to care for puppies than others. Sometimes you’d be better off adopting a dog that’s already been house trained. Or to pay for a trainer. Or to get a cat instead. Or a goldfish. Any resource that has a cost of upkeep needs a sustainability plan. Resources that do more will often—though certainly not always—require more initial investment and more upkeep. A puppy that is going to be a family pet requires a different level of investment—both up front and ongoing—than a rescue dog or an agility dog.

The utility of an open license, part 1

Of course, “good enough” and “better enough” is not the only way to frame the value of either affordability or OER. This is where the conversation among OER advocates can (and did) get chaotic quickly. For example, one topic that came up during Q&A was the utility of an open license to enable faculty to customize the content to their students’ needs. I said that faculty can and do customize with proprietary content all the time, and that anyone who believes the only way to do this is with OER is fooling themselves.

This comment caused some consternation. I stand by it.

When faculty want to adjust proprietary content, they skip chapters, supplement with other content (including some they may have made themselves), reorder the content, combine different texts, and so on. Instructors are endlessly creative in the ways that they slice and dice proprietary content. In fact, this exact tendency is one reason why textbook publishers got into pricing trouble in the first place. They have known forever that English comp professors are likely to use maybe one out of every three readings in an anthology, and that the particular readings which get used will vary from professor to professor. So they produce anthologies with three or four times as many readings as any class could use in a term. They do the same thing with problem sets. Or textbook chapters. Maybe one Biology 101 professor likes to spend more time on cellular biology while another is into ecology. No problem; the publishers just put in lots of chapters on both. Faculty will use what they want.

What you end up with by employing this publishing strategy is a puppy that has been house trained, agility trained, and rescue trained. That is one very expensive puppy. And the students—who, after all have to buy that puppy—get irritated because it is obvious to them that they will never have to use their dog for an avalanche rescue.

Customization happens regardless of license. Yes, a license makes certain kinds of customization easier. It’s an affordance. If you train your puppy yourself, you can decide what you want to teach it to do. Our dog, who we adopted as an adult, was trained by her prior family to ring a bell with her nose when she wants to go out. That was apparently useful to them. But I’ve had lots of dogs during the course of my life, and we never had to train them to do something specific when they wanted to let us know that they had to go out. Each had his or her own way of accomplishing this function. We didn’t suffer from that particular loss of control.

The definition of open education is an open question

One weird aspect of the “free as in puppy” analogy is that it treats the dog like a possession whose primary salient characteristics are cost of purchase and cost of ownership. That’s certainly one valid way to think about curricular materials (though not about puppies). But if what you’re really interested in is a pedagogical approach—let’s call it “open education”—then this is not the only way, or even the best way, to think about OER. Some OER advocates are interested in open education as a way of teaching, with OER being a set of raw materials designed to support that way of teaching. The problem is that we don’t have anything close to a consensus on what “open education” actually is.

Some of the summit attendees talked about the value of having students create and edit the content. Say you want your kid to learn some responsibility and empathy, as well as something about animal behavior and psychology. Having a dog will give them some of that. Having your kid train the dog will give them a lot more of it. Having two of your kids train the dog together will also teach them something about cooperation. In this case, the “cost” of training and caring for the dog is actually a benefit.

There is overwhelming evidence that having students learn by doing (including by researching and authoring) can be very effective. But there are two caveats regarding how this general principle of learning translates into the specific activity of student co-creation of curricular materials. First, having students write and edit their own curricular content is not inevitably effective as an active learning strategy. Sometimes, sure. But like everything else in education, it’s highly context-dependent. Second, depending on how broadly the students are sharing this work, it’s not clear that you need an open license on it, or that you need all content and source materials to be openly licensed. If the instructor doesn’t put any license at all on the student-created content but makes it freely available on the web, is it OER? In spirit, probably, but that would not be consistent with common usage of the term.

As a teaching strategy, I’m enthusiastic about having students co-create curricular materials. As a teaching philosophy, I’m agnostic and utilitarian about it. As teaching dogma—no pun intended—I’m deeply skeptical, as I am about all blanket generalizations about the “best” way to teach regardless of context.

My favorite variation on student production of curricular materials as a teaching strategy is Mike Caulfield’s notion of choral explanations. It adds the dimension that having a handful of different explanations can be more helpful than having just one. Think about your own behavior when you’re looking up a health condition on Google or a how-to demonstration on YouTube. Do you tend to look at just one search result? Or do you look at a few different ones? I often look at a few, and sometimes more than a few. The reason we can have the benefit of that diversity is because, on the web, there are many different people producing content resources and sharing them for free. They may not have Creative Commons licenses, but they are OER in a real sense.

Again, I’m enthusiastic about this approach as a teaching strategy and utilitarian about it as a teaching philosophy. If it works for your students, in your subject, with your pedagogical activities, that’s great.

Not all open education advocates define it this way. And to be clear, I’m not trying to provide a comprehensive list of useful definitions for open education here. I’m summarizing and reflecting on last week’s conversation. But I do want to touch on a subset of those other definitions that are often less well defined, more essentialist, or both. Because they did come up in that conversation, and because that’s where there are real problems.

Real problems

One question I got in the Q&A was what advice I had about things that the open education community isn’t doing as well as it could. That’s easy: Stop bickering so much.

Teaching, when done right, is deeply personal. That’s both good and bad. On the good side, first, many educators are motivated to be good teachers even when their environment actively disincentivizes them, because they care about their students. It’s one reason why educational systems produce so many success stories in spite of the fact that the systems themselves are deeply, disturbingly messed up. Also, there is evidence that the very fact that a teacher cares about a student has a strong chance of positively impacting that student’s physical, emotional, and financial wellbeing for the rest of their lives. (Many of us have our own personal stories about this, but there is also hard, longitudinal evidence.)

The bad side of teaching being so personal is that it can be very hard to maintain ego boundaries when you care so deeply about so many students in a messed up environment like the modern classroom (or, really, like human existence in general). I speak from personal experience as well as from knowing, working with, and living with many educators over the course of my life. It’s hard to separate the job from your personal identity. For some people, “open education” is an affinity group of sorts.

In and of itself, that’s fine. Teaching is hard, teachers deserve and need emotional support, and one way to get that support is to find your “tribe.” But I have observed a lot of infighting about shibboleths that mark membership in the open education tribe. Too often, it gets unreasonably heated and personal. At its worst, this behavior metastasizes into a particularly noxious form of identity politics. At that point, it is no longer about helping students.

As I said in answer to the question at the summit, there are only two essential goals that I care about in education: (1) increasing access and (2) increasing the value that students get from the education that they can access. I am agnostic and utilitarian about everything else. To the degree that discussion, debate, or usage of open education teaching strategies or open educational resources furthers one or both of those goals, then I’m for it. To the degree that it distracts from activities that could further those goals, then I’m against it. In recent years when I have attended conferences that are billed as “OER” or “open education” events, I have not been impressed with the ratio of constructive conversations to painful distractions.

This is absolutely fixable—if the participants decide that it is something they want to fix. I hope they do. Some of the brightest, most talented and dedicated educators I know are among these people. I would like to see them accomplish all the good in the world that they can.

My talk, and the utility of an open license, part 2

All of the conversation I described above is important. I’m glad we had it. But it wasn’t the conversation I had hoped to provoke. In the main part of my talk, I recapped the four levels of empirical education:

1. Intuitively empirical: This comes down to whether you pay attention to your students and do something differently with them based on what you observe. Do you always do the same thing, or do you have a bag of tricks that you can draw from when you see students struggle? I believe that the substantial majority of educators are empirical in this sense. They may not think of it as empiricism, but they are observing student behaviors and are adjusting their strategies based on what they see, guided by some sort of rationale for choosing which strategy to employ in different circumstances.
2. Mindfully empirical: Mindfully empirical educators think about how they can get the maximum amount of useful diagnostic information from day-to-day course work. They design their courses with a goal of creating many feedback loops that enable them to be adjust their teaching to the needs of the students.
3. Meta-cognitively empirical: Meta-cognitively empirical educators are empirical not only about how they use their existing bag of tricks but also about which tricks they should have and how effective those tricks really are. They consciously and regularly test their own assumptions about effective teaching, and they are open to trying new appropaches. My read of Lauren Herckis’ research is that the barrier of moving from mindfully empirical to meta-cognitively empirical (and to the next level, socially empirical) is where a lot of the difficult work needs to be done. Lots of educators are intuitively empirical, and the transition from there to mindfully empirical is not a huge leap. Getting them to test and challenge their deeply held beliefs about what constitutes effective teaching is a lot harder.
4. Socially empirical: Socially empirical educators view effective teaching not as an individual art but as a shared pool of knowledge and experience that everyone can learn from and contribute to. They seek out common vocabulary, methods, and standards of proof so that they can learn with their colleagues and raise the collective bar. This is the beginning of disciplinarity.

One possible defining purpose of “open education” is fostering socially empirical education. We can learn together and teach each other, teacher to teacher, teacher to student, student to student, and student to teacher. We can collectively learn how to teach and learn more effectively. We can conduct experiments, check each other’s work, and develop shared notions of what constitutes “evidence” of effective education. With this formulation of open education, as with the others, an open license is not a necessity; it’s an affordance. But substantial kinds of openness are essential to socially empirical education. You can’t build a shared body of knowledge without sharing.

The strategy of having students collaboratively construct knowledge artifacts fits in with this ethos nicely. Students learn how to negotiate the development of shared understanding. Here again, I am endorsing a strategy, not a dogma. But by framing student creation of curricular content as “learning how to negotiate the development of shared understanding,” we take the motivation for the educational activity out of the realm of say, social constructivism, which individual instructors may or may not buy into, and reframe it as a life skill that all humans should have. Doing this will help more instructors better understand why, when, and possibly even how they might want to utilize the strategy of having students co-create curricular materials.

Framing open education in terms of socially empirical education is intended to be a provocation rather than an argument. Since I am a critical friend of the open education community rather than a member of it, I don’t really get a vote. But I hope that the notion of socially empirical education can enrich the conversation among proponents of open education.