Saturday, February 18, 2023

The Cost of Free

At many colleges and universities there has for years now been a push to encourage the use of free OER materials as course texts. At not a few institutions, faculty are being not just encouraged to consider free OER materials, but pressured to go this route. This post looks at the background, and assesses the degree to which the push for academics and students to rely on free online learning materials may be harmful to learning.
The Price of Textbooks: It’s not hard to see why the idea of providing college and university course text materials free-of-charge became very attractive in the late twentieth- and early twenty-first centuries; many textbook publishers were setting absurdly high prices for their publications. Even in the late 1980s and early 1990s not a few textbook prices seemed to be driven purely by greed on the part of the publisher. Here’s one example: a bestselling textbook from that era, which was published in modest sized paperback format (providing an overview of its subject in fewer than 300 pages, with no color illustrations or “bells and whistles” that might add to the production cost), was priced for many years at roughly $100. The subject, ironically enough, was moral philosophy. Because the author was well-known and the book well-regarded, the publisher got away with it; a considerable number of academics kept adopting the book year after year, no matter the price, because they regarded it as the best book of its kind. But many other academics (and certainly many students) were outraged—and rightly so.

By the turn of the century, it was not uncommon to find many of the large multinational publishers pricing large, hardcover textbooks at $200 or $300. And in the 2010s the prices of the bound book versions of many university textbooks published by those large multinational publishers went even higher, as those publishers began to realign their approach so as to encourage book rentals and all-electronic delivery models. (As a byproduct, of course, the used-book market for course texts went into a steep decline.)

The Call for OER: It’s no wonder that more and more academics began to look for alternatives to traditional textbooks. One popular choice, as a sensible alternative to course text compilations that had done little more than bring together a selection of academic articles between two covers, was to abandon such textbooks in favor of posting the instructor’s own choice of articles (the rights for which are typically already cleared for university-wide use through the university library) on the institution’s learning management system. That’s an approach that continues to work well for many instructors in many courses. But there were and are a wide range of courses for which no collection of academic articles could possibly be an adequate stand-in for a textbook. Many people thus began to call for free textbooks to be developed—and that call began to be answered. The move towards OER (Open Educational Resources, as these free materials came to be known), steadily gathered steam. A number of state and provincial governments offered support, as did many foundations, and hundreds of academics began to make learning materials that they had developed available online, for free.

In some cases, these materials were electronic textbooks that professors had developed for their own courses but had been unable to find a commercial publisher for; in other cases, they were electronic textbooks by academics who had no desire to deal with a commercial publisher. And many bore little resemblance to the traditional textbook; some instructors posted point-form lecture notes to OER sites, and some developed sets of student exercises for use either with or without an accompanying textbook.

It should be emphasized that some OER resources were and are highly innovative in nature and absolutely first rate in quality. A high percentage of the best ones seem to be in the Pure and Applied Sciences, but there are numerous examples in the Humanities and Social Sciences too. In Philosophy, the for all X textbook, for example, is very highly thought of, as are some other OER resources in logic and critical thinking. OpenStax textbooks appear to be developed through a professional process of review and editing that resembles that of traditional publishers, and to be generally of fairly high quality. In English Studies, a good example of a high quality free online educational resource is surely the For Better for Verse prosody site developed by Herbert Tucker, a University of Virginia professor who has long been one of the world’s leading authorities on poetry of the Victorian period. The site accurately describes itself as providing
an interactive on-line tutorial that can train you to scan traditionally metered English poetry. Here you can get practice and instant feedback in one important way of analyzing, and developing an ear and a feel for, accentual-syllabic verse.
In other words, students can use the site to improve their skill at scanning poetry—a talent that has become relatively rare, but one that remains one of real value in understanding and appreciating the vast majority of all poems written in English between 1550 and 1900.

Not All Publishers Are Alike: The policies of the large multinational publishers that charge hundreds of dollars for a few months’ access to a single textbook get a good deal more attention than do the policies of smaller publishers such as Broadview Press. But Broadview is far from alone in its commitment to providing books and other learning materials of high quality at very reasonable prices. Other leading publishers in the Humanities and Social Sciences—among them Hackett Publishing (which for decades has been renowned for its commitment to fair pricing), W.W. Norton, and Oxford University Press—have similar principles. The prices charged by many large multinationals have risen far faster than inflation over the past several decades (as, it might be added, have post-secondary tuition costs for students—at least in the United States). But the trend has been in the other direction with publishers such as Norton and Hackett and Broadview and Oxford; the prices of books from those presses have typically lagged inflation over the past several decades; in real terms, they have typically edged downwards.

Professors and students still sometimes complain that the price of a large anthology from Norton or Hackett or Broadview is, at $60 or $70 or $80, too high. But these volumes typically run to over 1,000 pages and cost hundreds of thousands, even millions of dollars to develop. If Norton or Hackett or Broadview were to demand the sorts of profit margins that many larger publishing and media corporations look for, those anthology volumes would be priced at $200 or more. By comparison, prices of $60 or $70 or $80 for such large volumes represent extraordinary value.

But that’s $60 to $80 more than zero. Why should students pay $70 for, say, an anthology of American literature when they can get an American literature anthology for free?

The Example of American Literature: Let’s look at some examples of what you get for the price you pay. Compare, for example, the ways in which some of the ideas of Henry David Thoreau are summed up in The Norton Anthology of American Literature, The Broadview Anthology of American Literature, and a leading OER anthology of American literature:

By the middle of the twentieth century, Thoreau’s literary reputation equaled or surpassed Emerson’s with Walden regarded as one of the masterpieces of American literature. It is a work that powerfully affected environmentalists like John Muir and Aldo Leopold, who greatly admired Thoreau for his concerns about the natural habitat; contemporary writers like Edward Abbey and Annie Dillard also cite him frequently. As crucial as environmental concerns, and even anti-slavery, were to Thoreau, however, he always insisted that all principled action had to be undertaken by the individual rather than through groups. …

from the author introduction to Henry David Thoreau, The Norton Anthology of American Literature (edited by Robert S. Levine et al.)


Thoreau remains a vitally important reference point in discussions of the place of individual humans in human society as a whole, and of the place of humans (both individually and collectively) in the natural world. Walden—its status now secure as a foundational text of American nature writing and of twentieth- and twenty-first century environmental movements—continues to strike a sympathetic chord among a wide range of readers. Thoreau at one and the same time posits Nature as a magnet attracting all that is best in humanity and presents a compelling vision of Nature as standing apart from humanity—even in opposition to human society. “I love Nature partly because she is not man, but a retreat from him,” wrote Thoreau. “None of his institutions control or pervade her.”

from the author introduction to Henry David Thoreau, The Broadview Anthology of American Literature (edited by Derrick R. Spires et al.)

American Literature I (a leading OER American literature anthology):

Henry David Thoreau sought to live an essentialist life, one devoid of the unnatural excrescences loaded upon individuals by society and societal institutions. By realizing self-unity and being true to his individual self, he sought to realize his true selfhood as an organically-rendered microcosm of the macrocosm that is the world in nature. For Thoreau, nature has subjective value and meaning and shapes not only the body but also the mind and spirit. When such external institutions as the church and the government divert the individual from the overarching unity of themselves and nature, then Thoreau thought the individual should prefer integrity over conformity. Thoreau distills philosophical thought—such as Transcendentalism—and objective, sensory, scientific collection of concrete facts—such as Darwin claimed as his methodology—into a unique expression of integration: of self with nature, of self with culture, of culture with nature.

from the author introduction to Henry David Thoreau in American Literature I: An Anthology of Texts From Early America Through the Civil War, edited by Jenifer Kurtz), a free online anthology published with support from Affordable Learning Louisiana
It’s hard to know how to begin to unpack the muddle that we are offered in this passage from an OER textbook. And this is not a particularly egregious example; if anything, the introductions in the American Literature I text may be rather above average among the various OER anthologies of American literature.

If texts such as these become the norm, we will as a society have paid a very steep price. When one makes direct comparisons of this sort between the available OER learning materials and the learning materials that are so often denigrated by OER advocates—the learning materials provided by established publishers—the costs of free start to become clear. If academics are pressured into choosing OER with no concern for issues of quality—as happens with distressing frequency nowadays at many North American colleges and universities—the result will frequently be that students and faculty are denied access to high quality course texts. And when that happens there must surely be a domino effect; those who have had to make do with substandard learning materials in university courses in such subjects as English, History, and Philosophy will inevitably be at least to some extent less likely to achieve high levels of reading and writing skill themselves, less able to contribute at a high intellectual level in the world of work, less able to draw upon a large and well-constructed body of general knowledge, and less able to contribute intelligently and coherently to society as citizens.

There is no reason to be too hard on the authors of American Literature I, or on any of the other authors of the introductory material in this and other OER anthologies. As those at Broadview, Norton, Hackett, and other reputable Humanities publishers are aware, it is extraordinarily difficult to create a comprehensive anthology of consistently high quality that meets the pedagogical needs of students and professors. To do so requires years of work by a large team of academics and editors, and it requires a very large financial commitment as well; budgets for developing multi-volume anthology projects typically run will into the millions of dollars (investments that are typically not recouped for many years). Extensive editorial and market research is required. To ensure accuracy, introductory materials and annotations need to be rigorously reviewed not only by one or more of the main editorial team, but also by specialists in particular areas where the main editorial team may not have expertise. To ensure readability and consistency of style, everything needs to be carefully copy-edited and proofed repeatedly—by professional editors and proofreaders. There must be a substantial budget as well to pay copyright permission costs; even for anthologies focused on earlier periods, there are always materials that deserve to be included but that for one reason or another are still under copyright protection. With the best will in the world, it is almost impossible for a small group of academics, working on their own (without the many decades of experience that any professional publisher brings to the table, and without the help of professional researchers, editors, proofreaders, and designers) to create an anthology of high quality covering a subject as vast as all of American literature. With a project of much more limited focus, or with supplemental resources, leading academics such as Herbert Tucker have shown that it is possible to create truly outstanding learning materials that can be made available for free. But, as the example above illustrates all too well, it’s a tall order to replicate that sort of success with a textbook covering a topic as vast as all of American literature.

The Example of American History: A considerable number of OER options are available in History, but the quality of the material is for the most part highly suspect. Here, for example, is how slavery in early America is described in American History I: Colonial Period to Civil War, a college-level OER textbook by J. Franklin Williamson and Thomas Aiello, both of Gordon State College in Barnesville, Georgia:
...[A]s the profit motive for the colony was being borne increasingly by tobacco farming, settlers intensively sought after increasing amounts of farmland. By the late 1600s, “free” land in Virginia was becoming less and less available to these English colonists. Consequently, England attracted fewer indentured servants. At this point, slavery became more attractive and more desirable to the larger and wealthier tobacco planters. This was partly due to the need to grant land to emancipated indentured servants; it was also due to tensions between poor former indentured servants-turned-farmers and their wealthier former masters. With the increasing presence of Africans and African-Americans in the late 1600s, Virginia began passing laws that made hereditary slavery binding on all African-Americans in the colony. Those who had once served as indentured servants alongside whites, and could own land and even their own slaves once their contracts were up, gradually found themselves discriminated against. Increasingly, the African-American population in Virginia became slave-based.
Though the writing here is in need of a good editor (the authors presumably mean to say that the economy became “slave-based,” not that the population became “slave-based”), it’s certainly not as obviously muddled as that of the OER American Literature text discussed above. In this case, the more serious problem is the way in which the writing disguises agency. Slavery “became more attractive.” People “found themselves discriminated against.” The population “became slave based.” It’s written as if no actual human beings ever made a decision to enslave other human beings.

It's important here to note that the best American history survey textbooks are not exorbitantly priced. The options for buying an ebook of the first volume of the leading commercial American history textbook, Give Me Liberty, start at $30, and you can purchase a copy of the bound book for $62. For the full, two-volume set prices start at $45 for the two e-books, and just $75 for the two bound book volumes—this for large format, heavily illustrated volumes that total over 1,400 pages. Published by W.W. Norton, Give Me Liberty is written by Eric Foner of Columbia University, who has for decades remained one of the most distinguished of America’s historians and also one of its most lucid prose stylists. And the Foner is not the only option of extraordinarily high quality; it’s not even the only option of extraordinarily high quality from Norton, which also publishes Jill Lepore’s These Truths, a 960-page overview of American history that’s as lively as it is insightful, and that retails for just $19.95. That university administrators would be pressuring faculty to assign a free OER American history textbook for their students in order to save them the $19.95 they would spend for a copy of the Lepore, or the $30.00 they would spend for a copy of the relevant Foner volume, flies in the face of all the educational values that a university is supposed to stand for.

The Issue of Quality, and the Money Behind OER: In Williamson and Aiello’s defence, it’s worth pointing out that quite a number of students at Gordon State University may indeed find it hard to afford $20 or $30 for a high-quality American history textbook. Williamson and Aiello may well have been acting with the best intentions in the world towards their students; the average household income in Barnesville, Georgia is just $37,688. Their text, though, was subsidized by the Georgia government; the website was “created under an Affordable Learning Georgia G2C Pilot Grant.” The site does not say how much money was provided to the authors, but it’s public knowledge that numerous state and provincial governments in the Unites States and Canada have been encouraging such projects with large subsidies. Surely in this case the money would have been better spent on providing book vouchers to hard-up students (so that they could purchase the high quality $20 or $30 textbook) than on subsidizing the creation of free but substandard learning materials.

Where no learning materials of high quality exist, or where publishers are charging truly exorbitant prices, it no doubt makes good sense for governments to encourage the creation of OER learning resources through subsidies. But in the Humanities generally—and certainly in areas such as the American history survey course, where established publishers offer learning materials of outstanding quality at reasonable prices—governments would surely be far better off spending taxpayers’ money by providing students with book allowances—or by simply buying a copy of the Foner or of the Lepore for every student.

It’s also worth paying attention to the funding sources through which instructional resources are developed, and the potentially pervasive effects those sources may have on content. For traditional publishers such as Broadview, Hackett, and Norton, funding comes directly through sales: ultimately, a textbook is viable only if it fits the needs and preferences of the instructors who assign it and the students who buy it. With OER texts, on the other hand, funding is allocated up-front, without any accountability to the instructors or students for whom those works are intended. As a result, many OER texts are transparently lacking in fitness for the classroom, with little to no research done to structure content and no mechanism for tracking effectiveness or usage (and no incentive to do so). Some OER projects exhibit strong biases of a sort that is clearly aligned with the ideological orientation of their funders. The content of Business Ethics in a Box, for example (a recently-released OER website of resource materials for business ethics courses) shows a strong partiality toward free markets and against unions—biases that seem to align with those of the Templeton Foundation, which provided funding for the project.

Many proponents of OER use language that suggests that OER products, unlike the publications of commercial publishers, are free of the grubby taint of money. The Power Point presentation that Creative Commons provides in its efforts to make “The Case for OER” enthuses that “open licensing is the critical lever between modern education and the collaborative culture of the Internet.” And it’s true that the authors of many OER projects have prepared full course texts and made available online entirely without financial compensation; indeed, some OER advocates seem to recommend a “work for free” model of textbook creation (which may not have much appeal to instructors burdened with a 4/4 teaching load, or for academics whose teaching load may be lighter but may face onerous demands to produce research). The reality of OER nowadays, though, is that it’s often far from free of the taint of money—whether the money be grants from large foundations driven by political ideology or, in not a few cases, intrusive advertising. As an example of the latter, it’s worth looking at one more example: the OER site most often recommended by writing instructors in English Departments—the Purdue Online Writing Lab (universally known as “Purdue Owl”), on which it has become impossible to avoid the “partner content” inviting students to purchase Grammarly, or to “get a real writing expert to proofread your paper before you turn it in,” or to subscribe to “Citation Machine, a Chegg service.”

The Example of English Grammar and Usage Learning Materials: If you ask English instructors what they recommend as a reference guide to grammar and usage, the most frequent answer nowadays is “I send them to Purdue Owl.” And Purdue Owl is—despite the intrusive advertising—almost certainly the best of the free grammar and usage resources. But it’s very far from being one of the best grammar and usage resources. Compare the way in which Purdue Owl treats any given topic in grammar and usage that gives students real trouble (dangling modifiers, say, or the complicated issue of what makes a sentence grammatically complete) with the way in which the same topic is treated in any of the leading writing handbooks from respected Humanities publishers (A Writer’s Reference, say, or The Broadview Guide to Writing, or The Norton Field Guide to Writing). The treatment accorded each topic in the texts from the respected publishers is in every case more thorough and more accurate than that provided in Purdue Owl.

One might as well be blunt: Purdue Owl is often simply incorrect in what it tells students. “In English sentences,” it tells us as it tries to explain the issue of dangling modifiers, “the doer must be the subject of the main clause that follows.” The site’s authors are trying to explain what has gone wrong with hypothetical examples such as this one: “Having finished the assignment, the TV was turned on.” But neither the examples nor the advice is of much use in helping students recognize and avoid this sort of error in longer, more complex sentences of the sort in which such errors are usually made. And what Purdue Owl tells the student is also flat-out wrong as a generalization about the English language—as is demonstrated, ironically enough, in one of the examples of a correct sentence that’s provided by Purdue Owl itself a few lines further down:
They failed the experiment, not having studied the lab manual carefully.
Here the “doer” of the participle “not having studied” is “they”—the subject not of “the main clause that follows” (as Purdue Owl has told us must be the case in English sentences) but of the main clause that precedes the participle.

If all the good guides to grammar and usage from respected publishers were priced at $200 or $300—or even at $100—recommending Purdue Owl to students would arguably be the best way to go. But that’s simply not the case. A new paperback copy of The Norton Field Guide to Writing, purchased from the publisher, costs $58.43; ebook options start at $35.00. A new paperback copy of The Broadview Guide to Writing, purchased from the publisher, costs $44.95; ebook options start at $31.56. These are both reliable, comprehensive reference texts that run to roughly 800 pages.

The Research: What about the research? Sites that promote OER are awash with claims that “the research” has shown OER learning materials to be equal to or even superior to “traditional course materials.” Those who use Open Educational Resources, it is claimed, are likely to actually improve student performance compared with those who use textbooks from established publishers.

Dig a little deeper.

First of all, such research is overwhelmingly focused on courses in the pure and applied sciences—areas where the problem of exorbitantly priced textbooks has always been most acute, and areas in which (no doubt not by coincidence) available OER learning materials tend to be of far higher quality than they are in the Humanities and Social Sciences.

Second, much of the research is based on subjective reports—on students and instructors self-reporting, saying that they felt the OER learning materials were at least as good as learning materials that the students would have had to pay for.

Third, where the research uses objective criteria, those criteria are often strangely limited. Studies will conclude, for example, that OER learning outcomes were just as good on the basis of a similar percentage of students having achieved a passing grade, or a grade of C- or better. But what was the average grade? In most cases we are not told.

Fourth, the available research in almost all cases has been conducted by those already committed to the cause of promoting OER (and, in many cases, has been published in journals such as Open Praxis, the Journal of the International Council for Open and Distance Education, which are as devoted to the cause as are the authors). Given the degree to which educational administrators seem eager to find reasons to expand OER, it is perhaps surprising that there have been any truly open-minded studies published. But there have been some; one example is Regan E.R. Guring’s review article, “Open Educational Resources: What We Don't Know,” which concludes that “free, accessible curricular materials may have many benefits for students. But the research conducted so far hasn’t delivered the required proof.” Here’s an example of the sorts of observations Guring makes:
One of the first reviews of OER efficacy tests included 16 studies (Hilton, 2016). The abstract stated that “ … students generally achieve the same learning outcomes when OER are utilized.” If you stopped there you would be remiss. By the time you get to the results section, you note only nine of the 16 articles that made it into the review analyzed student learning outcomes. The other seven focused only on self-reported perceptions of the material.

All nine studies had major confounds such as method of instruction (e.g., comparing OER sections that were taught online or blended versus traditional texts used in a face-to-face class). Some studies switched exams between comparisons and some changed course design (e.g., went to a flipped model). Most study authors acknowledged that the type of textbook was not the only factor that changed. One study of the 16 minimized confounds by using a common pretest and exams and showed no statistically significant differences in use of an OER for chemistry (Allen, Guzman-Alvarez, Molinaro, & Larsen, 2015).

Astoundingly to the critical reader in me, many studies did not conduct statistical tests on differences (Hilton, 2016)…. Perhaps the biggest issue is that many large-scale studies of educational innovations do not use comparable exams.… In the one national multisite study using a common exam across classes, OER users did worse than traditional book users (Gurung, 2017).
It’s simply not true, then, to claim that “research” comparing free OER resources with textbooks from established publishers has conclusively shown OER textbooks to be of equal or better quality—shown (to use the language of the researchers) that OER textbooks “enable equal or improved learning outcomes.” No doubt more research is needed—particularly if it is conducted by researchers not already committed to promoting one side of the OER debate. But it does not take more research for an English professor to know that a textbook made up of sentences such as By realizing self-unity and being true to his individual self, he sought to realize his true selfhood as an organically-rendered microcosm of the macrocosm that is the world in nature is not conducive to learning. It does not take more research for a History professor to know that a textbook made up of sentences such as Increasingly, the African-American population in Virginia became slave-based is not conducive to learning. Academics should never be pressured by university administrators to choose such textbooks as learning materials for their students. At the post-secondary level in particular, the quality of textbooks and other learning materials is a matter of vital importance to learning itself; students will typically spend at least as much time reading their course texts as they spend in class. If textbooks and other learning materials are substandard, it makes it extraordinarily difficult for instructors to make up the difference.

The Future: Let me again make plain that I have no desire to demonize OER. Like Guring, I fully accept that free, accessible curricular materials may have many benefits for students. And there can be no doubt that many large multinational publishers have long made a practice of charging exorbitant prices for their products. But no one has demonstrated that free materials are inherently of higher quality than learning materials made available by traditional publishers. And—particularly in the Humanities and Social Sciences—there is abundant evidence that available OER materials are very often of far, far lower quality than are the course texts published by established publishing houses. There is abundant evidence too that—again, in the case of the Humanities and Social Sciences in particular—many of the course texts from established publishing houses are very reasonably priced. To spend large amounts of government funding in order to subsidize the creation and widespread use of inferior OER alternatives to these high quality, reasonably priced Humanities and Social Sciences course texts is, in effect, to waste taxpayers’ money, and to do a real disservice to students. A world in which governments and administrators continue to push for OER to replace traditional course texts, and in which the advocates of OER continue to ignore or downplay the issue of quality in Humanities and Social Sciences learning materials, will before too long become a world without the course texts that publishers such as W.W. Norton, Hackett Publishing, and Broadview Press have for many decades been proud to issue. That may seem like no great loss to those who persist in ignoring the cost of free. But the loss will be a very real one on several fronts. It will surely be a loss for scholars and for future scholars; the outstanding Humanities and Social Science scholars of today became outstanding scholars in part at least because they had access to outstanding learning materials. It will be an ongoing loss to instructors forced in course after course to cobble together learning materials from a range of deeply flawed free options. Most importantly, it will be a loss for students; a world in which students have far less access to high quality learning materials is a world in which educational opportunities are very substantially diminished.

What should be done to address these issues? Clearly, governments and university administrators should be taking into account the cost of free when they are shaping policies regarding OER. If the creation of OER learning materials is to continue to be encouraged, and subsidized, focus that encouragement and those subsidies on subject areas where the leading texts are priced exorbitantly, and/or on subject areas where there is a lack of high-quality learning materials—not on areas where high-quality options at reasonable cost are readily available. Take quality into account in all decisions as to the authors and projects that qualify for subsidies. Allow qualified publishers (those with a long track record of publishing high quality learning materials at reasonable prices) to become part of the process of creating, updating, and revising subsidized OER materials, lending their expertise in editorial and market research, copy editing, and design to the process. Apply mechanisms of assessment and accountability to government-funded OER projects, so as to ensure that such projects are of high quality and well-suited to course instruction.

Above all, stop encouraging academics to use inferior learning materials simply because those materials are free; don’t make students suffer by being forced to use inferior learning materials when much higher quality ones are readily available.

Full disclosure: Some of those who follow this blog are no doubt aware that I am the CEO and Company Founder of Broadview Press. I want to make very clear that this post expresses only my personal views, and not those of the company as a whole. I do not know how widely or how deeply the concerns I express about OER may be shared by others at Broadview; I do know, however, that all those in Broadview’s management group are persuaded that if any change on this front is be brought about, it is far more likely to be brought about by individuals (and in particular, by individual academics) drawing the attention of governments and of university administrators to the matter than it is by public statements issued under the imprimatur of publishing companies.

No comments:

Post a Comment