Justin Johansen and I did some interesting work on OCW sustainability, examining what happens when opportunities to enroll in for-credit courses are integrated into OCW. The results – over 2.5% of OCW visitors became paying for-credit customers of BYU Independent Study, generating enough revenue to more than pay for the cost of opening access to the courses. An article version of the dissertation, with a few months more data, is forthcoming in Educational Technology Research and Development.

John Hilton and I asked the question “What happens to printed book sales if digital versions are given away for free?” We then tracked 41 books for which we could identify the date when the free digital versions of the books were made available to determine whether the release of the free version affected print sales. This work appeared in the Journal of Electronic Publishing, and is described more extensively (with a link to the full text) at http://www.johnhiltoniii.org/the-short-term-influence-of-free-digital-versions-of-books-on-print-sales/. The results – three of the four categories of books had increased sales after the free books were distributed (and we think we know what happened with the fourth).

Next, John Hilton and I recently interviewed 10 authors (who were mostly academics) that have made their works available for free and got their take on the implications of sharing their books online. This article also includes a case study that measures sales of a book for the year before and after it was made available for free. An overview of the article, as well as a link to the full text, can be found at: http://www.johnhiltoniii.org/interviews-with-ten-authors-who-give-away-their-books/. The results – no authors perceived a drop in sales and most reported that giving away their ebook increased their reach / audience. This article appeared in Tech Trends.

Finally (for now), in a follow-up study that strengthened some weaknesses in the original, John Hilton completed his dissertation which also focused on giving away e-books. This study measured sales of 8 titles for 10 weeks before and after they were made available for free. Historical sales (from the year previous) as well as comparison books (that were not made available for free) are also examined. The study also includes download data for the books. An overview, as well as a link to the complete dissertation, can be found at: http://www.johnhiltoniii.org/free-e-books-dissertation-published/. The results – A 0.65 correlation between downloads and sales, meaning that books that were downloaded more often were also purchased in print more often.

John and I are currently working on an article based on the dissertation and a follow-up study suggested in the dissertation. We’re also looking for a medium to large size publisher who would be willing to do a replication of the dissertation research (with improvements for lessons learned, of course) at a larger scale. If you know of someone who might be game, let us know.

I’ve posted on my dissertation before. The full version is now available here. Here’s a little summary of what the dissertation is all about.

What

Deseret Book placed eight books online for free download. All of these were “backlist” titles. This study tracked what happened as a result of those books being available.

When

The books were placed online September 9, 2009. This study compares sales of these books the ten weeks before they were available for free with the ten weeks after.

Results

During the ten weeks of the study the books were downloaded 102,556 times. Collectively, the books sold 68 more copies in the ten weeks they were online for free versus the ten previous weeks. This was an increase in sales of 26%. Over the same period of time in 2008, sales of these same books decreased by 38%. Furthermore, a study of comparison titles that were not put online for free found that sales of comparison books decreased both in 2008 and 2009, as illustrated below.

	Featured Books	Comparison Books
2009	26%	-16%
2008	-38%	-6%

Thus the increase in sales of the eight featured books in 2009 seems attributable to their being available for free.

Other interesting findings include the following: Visits to the online product pages of the free books increased 1,085% during the study. Some weeks, hits to http://deseretbook.com/free represented almost 3% of total traffic to http://deseretbook.com. (note: the books are still available, but now require registration. They did not during the experimental period of the dissertation). During the ten weeks of the study more people entered http://deseretbook.com through http://deseretbook.com/free than any other page (except the home page). All this happened with very little advertising. Though the impact of this additional web traffic was not quantified, it seems that the value of increasing awareness of http://deseretbook.com could be significant.

There was a moderately strong correlation (r=.65) between downloads and Internet print sales (the more books that were downloaded, the more books were purchased online). Thus if more books had been available and downloaded the number of additional books sold would likely have increased.

Another interesting point is that the authors of the books made available seemed very pleased with the additional exposure their books received. One of the eight books studied was out of print. This book was downloaded 14,914 times and its product page received 834 hits. This may indicate a lingering interest in out-of-print books, and may validate a “long tail” approach to book sales.

Limitations

One limitation of the present study is the relatively small number of books studied, and that the sales of these books were relatively small. The study could also have been strengthened if it were of a longer duration.

One area that remains to be researched is the perspective of the authors whose books were downloaded. In informal conversations with the authors whose books were used in this study, they expressed positive feelings about their work being disseminated more widely. It may be that authors, particularly those with books that are not currently selling well, would desire the increased exposure to the work that free digital downloads might bring.

Although this study provides some interesting results, there are many other questions to be considered. One question that might arise is, “What would have happened if the e-books had been for sale for 99 cents each?” If 10% of the people who downloaded the books for free would have paid 99 cents for the e-books, that would have provided a nice profit for Deseret Book. While it is possible that some people would have paid 99 cents for the e-books, it seems likely that people would have been less inclined to e-mail friends and blog about a “99-cent” offer than a “free” offer.” Future studies are needed to determine how many downloads would occur if they were very inexpensive instead of free.

Another aspect that should be considered is how much the popularity of electronic book devices (such as the Kindle or iPad) will change people’s proclivities to read digital books. Perhaps most important, if publishers made all or most of their books available for free digital consumption, would that have an overall negative effect on sales? In other words, in the present study the seven books saw increased sales, likely due to the special attention they received from being downloaded 102,556 times. If all books were available for free online would they see this same sales benefit? Obviously not.

So, much more to think about and explore! Hopefully this dissertation will be a part of an ongoing conversation. They said nobody ever reads dissertations … prove them wrong … go read it!

First, the OCWC budget for 2010 is $1,000,000.

Second, the document includes a section called “Value Proposition to Members,” which includes the following explanation:

OCW Consortium currently offers educational institutions the following benefits:

OCWC membership…
• Provides an opportunity to fully participate in an international movement to increase access to education and knowledge, and to attract potential students to member institutions.
• Helps OCW proponents at member institutions in making a case for OpenCourseWare investment to university decision makers, funders and faculty
• Brings down the cost of creating and maintaining an OCW site
• Helps people all over the world learn about and find member institutions’ OCW websites courses
• Provides an opportunity to fully participate in the development of, and gain early knowledge of, toolkit innovations
• Provides the ability to participate in, and gain economies of scale in, the collaborative development of OCW Software
• Provides favorable publicity about member institutions’ role in the OCW movement and in the OCWC
• Provides a facilitated opportunity to learn best practices from other member institutions
• Provides first access to strategic alliances and collaborations with other member institutions
• Reduces the cost of participation in OCWC conferences for members

OCWC membership could…
• Support the generation of research funding at member institutions
• Support the rational management of course-related intellectual property at member institutions
• Attract funding for Open Courseware-related projects at member institutions with OCW sites
• Provide OCW-related technical infrastructure and support to member institutions

So now, with an overview of the value proposition, you can decide for yourself how good a value the OCWC is to the open education movement for $1,000,000 / year.

P.S. OCWC – please add this “Why You Should Join” list on your website where interested people can find it!

This dissertation is the first piece of empirical work I am aware of that demonstrates clearly that a distance learning program can simultaneously (1) provide a significant public good by publishing opencourseware and (2) be revenue positive while doing it. In other words, Justin’s study not only demonstrates that it is possible to publish opencourseware without requesting donations from users or foundations, it goes further and demonstrates that it is possible to make money publishing opencourseware. And if you can make money publishing opencourseware, you can continue publishing opencourseware for a very long time. This capacity is also known as sustainability.

Congratulations to Dr. Johansen!

P.S. John Hilton will be defending his dissertation, which applies a similar methodology to examine the impact of giving away free e-books on the sales of printed books, February 11!

In that post I invited people to send me more current information if they had it. Both Ryan and Steve Carson of MIT OCW accepted the invitation and provided more updated financial data. (In his reply, Steve good-heartedly suggested that I’m poor at math. My math was correct based on the numbers in Ryan’s article; I think Steve meant that I should redo my calculations based on more recent numbers.) So here we go.

In an article both Ryan and Steve suggested, called OpenCourseWare: Working Through Financial Challenges, we read:

To date, this effort has been funded by a combination of grant funding (41% of FY 2009 expenditures and 72% of total OCW expenditures since inception), Institute funds (49% in FY2009 and 22% of total to date), and donations and other revenue (10% in FY2009 and 6% of total to date).”

Using the FY2009 numbers, and the total of $3.6M for FY2009, that breaks down as follows:

MIT Internal funds – 49% – $1,764,000
Grants – 41% – $1,476,000
Donations, Amazon.com affiliate revenue, and all other sources – 10% – $360,000

This is a much better scenario for MIT OCW than the one suggested by the overall “since inception” numbers Ryan originally reported in The Tech. Still, 40% is a huge percentage of one’s budget, and MIT OCW is not out of the woods. The articles states:

In the next two years the grant funding that has supported OCW since its earliest stages will run out, and foundations generally do not provide new funding to support ongoing operations. Meanwhile, Institute funding has become tighter with the financial downturn, and like all units at MIT, OCW is under pressure to further reduce its reliance on the General Institute Budget. In the current economic climate, it is increasingly difficult to attract corporate support.

I didn’t realize this, but the article also states that MIT OCW has already cut its production activity in half – from publishing / updating 400 courses per year for five years to now publishing / updating only 200 courses per year.

However, none of the above items are the most interesting thing contained in the report, authored by Cecilia d’Oliveira and Steven Lerman. The most interesting thing of all (to me) was the following:

Proposals for generating revenue based on OCW are also reflected in the Institute-wide Planning Task Force Report [unfortunately, "You must log in to view the report."]. These ideas include various types of certificate, credit, or degree-granting distance education programs that rely on the OCW materials. At this writing, a pro bono team from management consultants Bain & Company is helping us assess the Working Group’s ideas in terms of their potential for financial return, alignment with OCW’s core principles as well as the perceptions of OCW’s stakeholders and users, and the cost of implementing those ideas.

As you may recall, this past May I wrote a post on The Future of OCW in which I forecasted that:

Every OCW initiative at a university that does not offer distance courses for credit will be dead by the end of calendar 2012.

BYU has already demonstrated a profitable “Click to Enroll” business model where OCW materials are used simultaneously as (1) a public good, and (2) marketing materials for credit-bearing programs. There is anecdotal (but positive) evidence from the Click to Enroll experiments at UC Irvine, the OU UK, and the OU NL (these are described in the BYU study). Reducing the cost of OCW development isn’t enough to make the program sustainable – I still contend that the OCW program has to generate real revenue in order to be sustainable.

If MIT OCW is going to adopt a “Click to Enroll” business model, then chances are that we can stop worrying about its sustainability. In fact, if MIT OCW goes C2E, they can probably become entirely financially self-sustaining, without relying on donations or the institution. They already take the lion’s share of OCW traffic, and despite the nature of their course offerings (e.g., the lowest math they offer appears to be single variable calculus) MIT OCW will likely take the lion’s share of the OCW Click to Enroll revenue, too.

The BYU study found that over 2.6% of all OCW site visitors chose the Click to Enroll option and became paying customers enrolled in formal online courses. Can you imagine the revenue generated by applying BYU’s conversion ratio to MIT OCW’s million visits per month? With a conservative tuition of $400 per class, that’s revenue of over $10,000,000 per MONTH. Even with a terribly low conversion rate of only 0.1%, they would still generate almost $5,000,000 in revenue per year.

As always we’ll all be watching MIT OCW closely and hoping for their success. As the article says, “More than 250 universities have committed to openly publishing course content in the OCW model.” Hopefully that will soon mean the Click to Enroll model, and OCW will become a sustainable fixture at all universities.

Since its creation, 22 percent of OCW’s expenditures have been covered by the Institute, 72 percent has been paid for through grants from the William and Flora Hewlett Foundation and the Andrew Mellon Foundation, and 6 percent has been covered by donations, revenue, and other sources.

(His article states that these numbers are “since it’s creation,” but they’re the best breakdown of numbers I know of. If you know of a similar breakdown for MIT OCW’s 2009 finances, please drop a link in the comments below.)

If we work these numbers out, each year that’s roughly:

- $2,952,000 for 72% covered by Hewlett and Mellon,
- $902,000 for 22% covered by MIT internally, and
- $246,000 for 6% covered by donations, corporate sponsors, Amazon.com affiliate revenue, and all other sources of revenue.

Ryan’s article is an extended argument for why MIT should continue to support OCW after its grant funding runs out in two years. I (and I expect most readers of this blog) agree with the importance he places on the project and the very important public good it has become. More importantly, MIT OCW is terribly important to the broader field of open education.

Because MIT OCW receives such a large percentage of the OCW world’s traffic and media attention, potential problems for MIT OCW are potential problems for all of us.

I keep asking myself how you support a project when 3/4 of its funding is pulled out from under it. Two years is not that far away. And it already feels like I’m getting a “Please remember to donate to MIT OCW” email once a month. On 25% annual budget, what would MIT OCW do? If MIT OCW were to go into stasis (like USU OCW recently did), how would that be viewed by the world?

More importantly, what is Plan B for the broader OER field? Imagine that two years from now MIT OCW announces drastic cutbacks (or temporary suspension) of its program. How do the rest of us argue for open sharing on our campuses then? Perhaps these arguments would revolve around the sharing model, or the way sharing happens – “we’ll do it differently in the following way…” Perhaps they would revolve around business models and using OCW to generate revenue (e.g., by using them to market for-credit online courses). How else do we make the argument for open sharing on our campuses in a post-MIT OCW world?

I’ve already heard “just because MIT can do it doesn’t mean we can” about a thousand times from faculty and administrators. What if that becomes “Not even MIT can do it for longer than a few years…”

My point is that openness is the virtue to be pursued (I feel silly making this statement to you – you’ve done more for this “movement” than almost anyone else has). Not sorta-openness. Or sorta-affordable openness. Full openness to download, edit, reuse, add media, etc. is the target. Settling for affordable quasi-openness may sell cheaper textbooks and may delay more foundational change.

I’m not sure how content licensed By-NC-SA can be considered sorta-open, unless we’re heading down the path toward an NC discussion (let’s please not go there). Full openness to download, edit, reuse, add media, etc. is what people have with FWK books because that’s what the license provides. What is the criticism here?

What are our organizational models missing when individuals are not capable of collaborating in writing complex resources (like textbooks)? Is it about incentives?

At least partly, if not primarily, yes.

Could we have networks of educators write textbooks?

With the proper incentives, of course we could. If we paid each faculty member $10,000 I bet we could get them to participate.

Let’s say a group of psychologist profs got together and decided to write a full text for first year students and posted resources in a wiki.

But – and I’m not trying to be thick-headed here – wouldn’t the size of the group of profs need to be rather large before this activity became interesting to you? When you say ‘a group of psychology profs’ I think of a group of 5 plus or minus a few. Is that a network? How is it different from five authors who develop a By-NC-SA textbook for FWK? In theory, one difference might be that anyone outside the core group could contribute to the text, but in practice we see that participation in “wikibook” projects from outside the core group is basically nonexistent.

If FWK is trying “trying to disrupt the status quo as much as we can as quickly as we can” why not experiment in serving as an infrastructure role in openness at this self-organizing level with profs?

(A) Because Wikieducator, Wikibooks, PBWorks, and a hundred other sites already provide the infrastructure necessary for the kind of experiments you’re talking about. (B) Because all the data to date confirm that those experiments fail to result in quality textbooks. In other words, the world doesn’t need FWK to enable these experiments – they’re already sufficiently enabled and they’re already failing to succeed.

Should we still be thinking in textbooks? What is it about textbooks that is so valuable that we transition the concept fully into the digital world? Maybe we should first abandon the textbook model.

We don’t get a clean slate when thinking about changing a system – we have to respond to the realities of the system. Some in the open education community seem to be arriving at a conclusion along the lines of, “the constraints of higher education are too confining, so let’s disconnect ourselves from HE completely and go innovate over there.” I wish them luck.

Does FWK permit one student in a class to download a book and then distribute copies to other classmates without fee? Can an educator download the book, copy and paste into a wiki and then edit it to customize the text?

Yes, full stop. The license FWK uses for all of its books, CC By-NC-SA, allows anyone and everyone to download, adapt, and redistribute the book, students and educators alike.

Collaboratively produced resources, in the wikieducator sense, are better suited for reuse because, in theory at least, no one has a motive – such as profit – other than to produce learning resources… Even the small reading window for reading a text online is an illustration of control exerted to influence purchasing the text. If a group/network collaborates on the text, then (again) in theory, they wouldn’t need to play “soft control” games of this nature.

There’s a lot of “in theory” in this line of thought. (As Firefly’s Jane once said, “I smell a lot of ‘if’ coming off of this plan.”) “In theory” is difficult for me to accept given the rather large amount of actual data and experience available to us to base our judgments on.

Pure openness would be the target.

I actually just finished blogging about this in Feeling Out of Place.

In my view, if it (research, course content) comes from the public purse it belongs to the public. If it’s privately funded, it’s a different matter. FWK is a private entity that is in business to make a profit. Nothing in the world wrong with that. But is it the best way for academics to approach opening up content/curricular resources?

It depends on their goal. To me, openness is a means not an end. So, to answer your question, we must ask what is the faculty member’s goal in being open? If “being open” has become the goal in and unto itself, I would propose that there is a problem.

David, whether you lay claim to the title or not, you are the (or at least “an”) ideological leader of openness in education. Which is why I was a bit surprised to see you accept the FWK model. I’m sure there are considerations I’m not aware of, or philosophical views that are perhaps not as strident as might be expected from a leading figure. To me, it seems to be trying to balance openness with economics… and the economic model has precedence (i.e. charges for downloads of digital versions).

One of the greatest heartbreaks of my life came when I left USU to come to BYU. For a complexity of reasons, some of which were my own fault, when I left USU the grant funding for many of the “open” projects at USU ended. Within 30 days of my coming to BYU, much of what I had worked to build over a 5 year period was gone. It was personally devastating. I committed to myself then that I would never again waste my effort on projects that can disappear overnight when grant money dries up. This has led me to adopt a keen focus on the sustainability of open education projects (see, for example, the study on making OCW pay for itself we’re conducting at BYU). If open education is to have a long-term impact we have to insure that it will survive over time. So, when you think you see an emphasis on economics in my work you are quite perceptive and absoultely right. However, I’m only interested in the sustainability of open projects – I have no interest in the sustainability of pseudo-open projects.

When Jeff and Eric first approached me about being involved in FWK (I’m not a founder, but am hire number one), my very first thoughts were “Are these guys going to do it right? Are they going to (1) get the openness right and (2) be sustainable enough to make a large-scale, long-lasting difference? Or are they going to take some half-open approach and/or botch the business part of it and be gone three years from now?” I was convinced FWK was going to get the business part of it right, and could sense that they wanted to get the openness part right. That’s why I joined as Chief Openness Officer – my core function is to make sure we get the openness part right. And because FWK produces Creative Commons licensed textbooks that are just as open as any content you will find anywhere, and because we have 40,000 students lined up for fall, I think the openness and the sustainability aspects of our work are going awesomely.

Your work around advancing openness, by nature of this role, will be subject to scrutiny. If you have a view on copyright or commercial reuse, it will be criticized. If you have a view on how to increase the impact of openness at the school or university level, it will be scrutinized. Is it fair? No. But that’s a burden that comes with the role.

No, the increased scrutiny is not fair at all – it is a great blessing that very few people have access to. It is a manifestation of the classic problem of the “rich getting richer” – I think I have some reasonable ideas on the topic of openness and education. These draw scrutiny, which I weigh seriously. Then my reasonable ideas get a little better, which gets them pushed out further, which draws further scrutiny, etc. Its a virtuous cycle that I am deeply grateful for; I actually feel guilty sometimes that other people’s ideas don’t get the “airtime” and the scrutiny that mine do.

And I’m enjoying this specific conversation that we’re having quite a bit.

In a blog post titled The Future of Newspapers, Posner opines that the best solution to the newspaper industry’s problem may be expanding the scope of copyright law:

Expanding copyright law to bar online access to copyrighted materials without the copyright holder’s consent, or to bar linking to or paraphrasing copyrighted materials without the copyright holder’s consent, might be necessary to keep free riding on content financed by online newspapers from so impairing the incentive to create costly news-gathering operations that news services like Reuters and the Associated Press would become the only professional, nongovernmental sources of news and opinion.

Now, I’m certainly not one of the country’s most respected legal scholars, but here’s some advice for the newspapers: IF YOU DON’T WANT PEOPLE LINKING TO YOUR CONTENT, DON’T POST IT ON THE WEB. That’ll be $2500 / hour for legal consulting, please.

On the other hand, there may be something useful hidden in this recommendation. Imagine momentarily that the Web had turned into a place where you could only link to pages whose rights holders had given you explicit consent to do so. The best mechanism for giving this kind of consent is, of course, the Creative Commons licenses. This proposal could go a long way toward eliminating links to fully copyrighted content, effectively eliminating it from the network (consider – if a writer posts a story in a forest but literally no one links to it, does it exist? Google and Yahoo can only crawl pages that someone links to) and leaving only a huge interconnected graph of CC-licensed material.

Not to mention the fact that all material is copyrighted, meaning that you wouldn’t be allowed to link to anything without the owner’s previous permission. And if you couldn’t link to it to look at it, how would you know whether you wanted to link to it?

If anything, this blog post shows that Posner understands nothing about the Internet. How embarrassing for him! Perhaps I needed to ask for his permission before linking to and commenting on his post?