I received a free copy in the mail of an introductory statistics textbook; I guess the publisher wants me to adopt it for my courses. The book isn’t bad, actually it’s pretty good: it follows the “Moore and McCabe” format, starting with descriptive statistics (up to correlation and regression), then a bit on data collection, then probability, then statistical inference, and at the end chapters on various more advanced topics.
I showed the book to Yu-Sung and he said: Wow, it’s pretty fancy. I bet it costs $150. I didn’t believe him, but we checked on Amazon and lo! it really does retail for that much. What the . . . ? I asked around and, indeed, it’s commonplace for students to pay well over $100 for introductory textbooks.
Well. I’m planning to write an introductory textbook of my own and I’d like to charge $10 for it. Maybe this isn’t possible, but I think $40 should be doable. And why would anybody require their students to pay $150 for a statistics book when something better is available at less than 1/3 the price?
This won’t be easy, because I’m planning to write an entirely new kind of intro book, starting from scratch. But why hasn’t someone written a more conventional book at a cut-rate price? Or maybe they have, and I just haven’t heard about it?
It just mystifies me that, in all these different fields, it’s considered acceptable to charge $150 for a textbook. I’d think that all you need is one cartel-breaker in each field and all the prices would come tumbling down. But apparently not. I just don’t understand.
P.S. More thoughts here.
That is one of my pet peeves too. Would like to see a cost breakdown and understand who's getting a share of the $150.
Also, have there been studies that show the value of a particular textbook on a student's grasp of statistics? That would surely make for some good reading.
Here in Canada, when universities post helpful "what you can expect to spend" budgets for prospective students, they include about $600 per semester for textbooks. At 5 courses per semester, and with some books surely covering more than one semester, that's a healthy chunk of change. So your $150 figure does seem to be "normal" nowadays.
If you want to control the price, I assume you're going to publish it yourself? Check out lightningsource.com. Very well set-up outfit, very good deal for author/publishers. You don't have to sell very many to make *lots* more than you get in royalties…
How is it in the teacher's interest to use a cheap book? I mean, sure, if the students could decide, we would all be using free texts from the web now.
Two priors come to mind. First, 37signals success with their "Getting real" .pdf.
Second, the "Pragmatic CSO" by Mike Rothman.
In both cases, a .pdf provided proof of concept and initial sales that then translated into independent (physical) publishing.
You're biggest pushback is going to be coming up against the traditional distribution system for textbooks rather than the ability to self-publish.
But I'd buy it…
Its the publishers. They try to sell these books with expensive paper, colored pictures, hardcovers…etc. in addition to the gimmicks like cd's, study manuals, websites…etc. These add nothing to the quality of the book and so much to the cost. Publishers have figured that most students are gullible and affluent enough to shell out the dow.
I'm so glad you're going against the tide! Please stay with CUP or some other reasonable publisher who will not try to screw us all over!
This paper proposes numerous reasons for the high cost of textbooks:
"AN ECONOMIC ANALYIS OF TEXTBOOK PRICING"
http://www.ed.gov/about/bdscomm/list/acsfa/kochreport.pd...
My theory has long been that the people choosing the textbooks are not the people who have to pay for the textbooks, so they're not particularly sensitive to price. Only some professors are concerned about the prices their students will pay for textbooks, while all students (or their parents) are concerned about the prices they're paying. This could be tested: are the prices of research monographs, which are often bought by the researchers who are using them, more reasonable? (But I'm not sure what "more reasonable" would mean in this case, since it's such a different market.)
What about using a self publisher like Lulu or Author Solutions? You would lose the marketing channel, but you have your blog!
Maybe the writer/publisher model should just be circumvented. In the last two stats classes I took, my study group took turns and scanned the whole book and distributed the pdfs, then we returned it to the library. =)
Someone should release an entire stats book in pdf format and use the Radiohead-model for payment. The author would get all the proceeds. I'd give you at least $10 for a brand new stats book in pdf. I really enjoy the ARM book.
I was always under the impression that the publishers drove the prices. I assumed they could get away with it because once a prof chooses a text book, the demand is almost totally inelastic. As you mention, there is probably some price sensitivity at the professor level, but I suspect most professors do not even look at the price when choosing a text.
I think that targeting a $40 intro text is a fantastic idea. You probably could get the price to $10 using an ebook (pdf, etc) model. But it seems really hard to get small runs of text on physical paper for under $40.
Once all students are carrying a Kindle or kindle-esque device, I think the pricing model for text books will HAVE to change. We shall see!
I'll look forward to you writing such a book. I'd probably buy it. My intro to social statistics textbook was 95 bucks USED. USED.
Textbook prices are insane anymore and amazon is quite useful.
That would be amazingly kind of you. I assume that most people charge $100+ for it because either
(1) they want to make money or
(2) their publishers make them so that THEY make money
I'd suppose it's because so few of the textbooks get bought, relatively speaking? I don't know. There are a lot of open (free) textbooks available online in pdf format. It would be useful if you could do that, too…
Textbook prices long have been a classic market failure. Professors don't care about price; you're a perfect example. Price isn't on your radar. Students can't care about price, since the book is needed to take the class. Nobody else will assign your $10 text to save their students money, though a few might if they like it and pleasantly surprise a few pocketbooks. And nobody wants to write a $10 text, since the publisher won't touch it and since every professor who writes a widely adopted intro text seems to end up getting rich off it.
The answer, it seems to me, is for enough educators who are offended at these practices to write intro texts gratis and post the PDFs online with some sort of creative-commons license. The good ones will be adopted, assigned, and run off for $10 by local copy shops or even by a small local publisher. Big publishers will be out of the rapacious intro-textbook business. And a few professors will have to subsist on tenure and salary.
Next on the hit list: journal prices.
This is indeed pretty common for text books. When I look at the price of the text books I use for my classes, it might even be in the low end of the scale… it seems the more specialised the topic is, the more expensive the books become.
I guess this makes sense. The more specialised the topic, the fewer sales, so the price goes up.
Whether it is reasonable is another matter. I have no idea about the price it actually costs to produce a good text book.
Increasing the price where there are fewer sales only really makes sense for the up-front cost of producing the book — the cost of actually writing the book, editing, proof-reading, setting up the text etc. — but less so for the per-copy costs of printing and distributing it.
It is a mystery, but I will hazard a few conjectures. A lot of people think that something that costs more must of necessity be better than something that costs less. But that doesn't explain why a professor who should know his field would choose a $150 textbook over an equivalent and cheaper alternative. Unless of course he's the author.
Not to be snarky, but just as you are mystified by the price of stats textbooks, I'm mystified that a professor who teaches regularly could not know how expensive intro textbooks are.
You should check out some of the physics and chem books.
It comes down to Copyright + Rent Seeking (in the econ-lingo).
There is a massive arbitrage market on the internet in which people buy textbooks overseas and import them to the US. typically abebooks.com and amazon.com have these available.
Typically they cost 50 to 70% of the new price, even with overnight shipping from Taiwan or whatever. If you get the "international editions" which are printed on super-thin paper with paperback covers they might cost 30% of the hardback US edition.
I think you should strongly consider offering your intro text as a PDF on the web free, in addition to your print version. The big advantage of having PDF versions is that you can search them.
That would be great! But even when teachers write their notes up and send them to the local copy place, the books still range around $30-$50. What about a totally electronic version? No print versions? You could sell licenses for the $40 that allows students to download a copy.
Do writers set the price of textbooks?
I'd guess it's because professors don't pay that much attention to the price of the book they're using, so books don't really compete on price. Will your book be Bayesian?
It's actually a very "american" thing to charge these ridiculous prices and constantly publish a new edition so that the second hand copies get outdated. In German the most comprehensive introductory stats book (there are not a lot) costs 50 euro (very nice hardcover, tons of color graphs) and I know professors who don't want to assign this book because it is to expensive…
As an undergrad, a couple of my professors provided us with textbooks they had written in PDF format, so it was free, albeit not as convenient as a bound book, but if we wanted we could have the PDF printed and bound at a local copy shop for cheap.
In my experience though, most professors did not seem to care how burdensome it was for us to buy the books.
As I see it, the publisher provides some publicity and a lot of credibility. The latter is what keeps cheap books away from most classes. With your reputation, you don't need that, and you could totally pull off a cheap book. I think the way to do it would be to self-publish via cafepress or somewhere similar. You'll probably get as much per book as you would with Wiley, too.
Not to offer any discouragment but – you may wish (if not already) to have a look at the teaching manual for Freedman, Pisani, and Purves. "Statistics: Fourth Edition." W. W. Norton, 2007
they apparently tried very hard for over a period of 10 or more years to get it "right" or perhaps better put "less and less wrong".
Now, in my opinion, they accepted getting across the sampling distribution of sums/means as one of the central tasks and the new addition has no modern computation nor Bayesian stuff in it at all – so lots of reasons to believe one could do better …
I can explain (but not condone it) as simple price elasticity. Textbooks, once chosen by an instructor, are highly price inelastic for the student, and as such, the publisher will raise its revenues by charging a higher price.
As for a textbook, I'm contemplating using 'Statistics in a Nutshell' by Sarah Boslaugh and Dr. Paul A. Watters as the suggested course textbook. It's more of a reference type book, but it's cheap (< $40), and contains most of the material of an intro stats course presented in a clear fashion. Coupled with good notes (and a prof who has read your 'Teaching Statistics' book), I believe it would a good choice. In addition, at that price, the student is likely to keep it, and refer to it in the future.
Andrew, I'm delighted that you're planning on writing an introductory statistics textbook. As a software guy who has learned a lot from your blog and the ARM book — both of which are not only informative but also fun reads — I would love to see your take on introductory statistics. That you're also planning on taking a stand against absurd textbook prices makes me doubly happy.
Keep on doing what you're doing.
Cheers,
Tom Moertel
If the professors choosing the books are not sensitive to price and the students have to buy the book, there's no motivation to sell texts for less. Profs are highly motivated to choose based on quality and appropriateness.
As a percentage of tuition, text costs are almost negligible (Columbia annual tuition = $35K, so that's well over $3K/class.) Couldn't one Ivy League University break the coalition and charge less?
I have two intro math stats texts: Larsen and Marx ($140 marked down to $110); DeGroot and Schervish ($70 marked down to $60). But the choice is likely to be made on the basis of DeGroot and Schervish being enhanced/polluted by Bayesian notions.
Have you considered making it open source? For example: http://www.opensourcetext.org. I've wondered about trying to persuade a publisher to do two versions – an open source version, and a printed version, and see if people will pay to get a nicer version. (I would, but I'm not a poverty stricken student). I have a vague idea of doing this with a book on R. (Of course, it's not a new model – Zen and the art of the internet did it before we'd even heard of this WWW stuff).
A few years ago I wrote a short text aimed at psychology students in the UK, and selling for 10 pounds (about $15, today). It didn't do especially well, and now it's out of print, so I turned it into a wiki, at http://www.researchmethodsinpsychology.com. No one reads it, but at least it's free. :) (Well, to them, I have to pay hosting, but bandwidth is pretty low, so that's cheap.)
Two comments to Bob Carpenter: Most students pay less than that for tuition, and a lot get scholarships. Even if they didn't, textbooks feel like marginal costs (even if they're not) when tuition fees feel like fixed costs. Maybe the way to do it is to include the textbook on the course – it's a small (percentage-wise) increase in fees, and the university (and presumably the professor) has an interest in using a cheaper book.
I can't believe that high prices don't hurt sales. There are two books by Lee that I would like to get: Structural Equation Modelling: A Bayesian Approach and Handbook of Latent Variable and Related Models, but there's no way I'm paying over $100 for each of them. If they were cheaper, I'd get them, or if they ever appear used, I'll get them – which means that the author, and the publisher, get nothing.
Finally, I look forward to seeing the new book – whatever form it takes.
Since you're on that topic, have you checked out the price for Teaching Statistics: a bag of tricks at Book Culture?
Teaching Statistics at Book Culture
Maybe the weight explains it all! More seriously, from an economic point of view, you would think the opposite would happen: many customers meaning that the marginal cost is negligible and many competitors meaning that you should sell at a lower price (assuming the quality is even). Since this does not happen, it means it is not a free market! First, publishers are few and can agree on high prices. Second, students are mostly captive of the book chosen by their professor.
What a wonderful contribution to future statistics students everywhere that would be.
My understanding about the tuition issue raised by Mr. Carpenter is that schools are afraid that a low price is a signal for low quality. Something like the old adage about how we don't mind that our competitors have lower prices, they know what their product is worth.
I really appreciate that Gelman's Data Analysis Using Regression and Multilevel/Hierarchical Models was an inexpensive softcover.If it had been another $100+ hardcover I've have skipped buying it.
There are some inexpensive Dover Books on Mathematics/Science that discuss introductory statistics but I think they are over 30 years old.
I was under the impression that the publishers set the price, not the authors. I wonder if there is grant money available to write public domain texts; that way, you could release it for free, and perhaps rely on print-on-demand services for students to get hard copies of the text. I hope you DO write an intro text though. Moore and McCabe is nice, but I think there's plenty of room for improvement.
I have often thought that if I ever finish my textbook (in law, not stats), this is precisely what I would do. While I can see charging a higher price for books that are used by professionals, I don't see why student books should be so expensive. It is not as if authors see much of that markup in their royalty checks anyway.
Self-publishing companies like lulu.com make Andrew's suggestion of $40 quite possible. In my experience, their quality is perfectly reasonable, and if authors want some compensation for their time and effort, $40 can become $60, but that's still not $150.
Michael Lavine is writing a book and just putting it up on the web for free download.
http://www.math.umass.edu/~lavine/Book/book.html
Moore & McCabe is not just a textbook. An instructor using it has access to a solutions manual, Powerpoint slides, and an archive of test questions; it can also come bundled with Minitab. I think even more support material has congealed onto the book, but I've lost track. A salesperson recently tried to sell me on an introductory statistics book, and was surprised when I asked to see the book, as if the contents of the actual textbook were irrelevant to whether anyone adopted it.
I'm currently poring over Boslaugh and Waters' Statistics in a Nutshell, O'Reilly Media, $34.99, to try and break the $100 barrier.
My experience is that professors' metrics of textbook quality for introductory statistics are the number of homework problems and the availability of a solution set and test bank. THEN they worry about content and organization.
Perhaps there needs to be some sort of "PLoS Textbooks" equivalent, whereby high-quality, reviewed textbooks could be distributed as searchable pdf files. There would then be no justification for $100+ pricetags, and the format would probably be more useful to the student.
I've been meaning to look at Collaborative Statistics by Susan Dean and Barbara Illowsky. Maybe it will interest you, too. It is an "open-access" introductory statistics book, which is available free as a pdf, or as a hardcopy from a print-on-demand service. Since it has been licensed as Creative Commons, even if you don't like it, you could possibly use portions in your own book.
I agree with Gregor in that I have the impression this is chiefly a US phenomenon. I think part of the reason is Mike Anderson's point that professors' metrics of textbook quality are the number of homework problems and the availability of a solution set and test bank. In other words, they're buying an entire prewritten course from the publishers – or rather choosing one and expecting the students to pay for it. In the UK we'd expect the lecturer to write the course materials and provide them to the students, so in many or even most cases buying a textbook is optional. UK-written textbooks are quite a bit cheaper, but have typically fewer or no problems and usually no solution manuals or test banks. If a textbook is required for a course, students at a decent university will expect the university library to hold many copies, some of them available on short loan only or confined to the library.
The same textbook is often cheaper in the UK. E.g. amazon.com says list price of 2008 hardback edition of Woodridge's Introductory Econometrics (with Economic Applications, Data Sets, Student Solutions Manual Printed Access Card) is $171.95, though they're currently offering it for $129.85. The "International Edition" paperback available in the UK (specifically "Not for sale in the US") is available new from three sellers on on amazon.co.uk marketplace for £46-55. Currently £1 = $1.44 but even the peak last year of $2 to the pound wouldn't explain the difference. I'm sure it reflects the different markets and different price elasticities – in the US, most students are required to buy a particular textbook, while in the UK, most can choose which to buy, if any.
I recall doing one mathematics subject (here in Australia, but a couple of decades ago now) where the lecturers took great care to choose two inexpensive texts for the subject. One was a very thin paperback from the UK and the other was a thinnish hardback (on very thin purplish paper) from Russia. Together they cost half what I was paying for a single text in most of my other subjects. They were also both quite good.
It made a big difference to my budget at the time.
[in passing–got here from CT]
I went through this, recently, advising colleages on a specialized monograph that was previously published through Booksurge. When Amazon took them over, Booksurge lost–one can't make this up–the book from their catalog. So I did some market research & here's what I found:
1. There are three main POD channels: Booksurge, Lightning Source, and Lulu.
2. If you want your book sold through Amazon, you have to do an edition with Booksurge.
3. If you want your book available in bookstores, you have to do an edition with Lightning Source.
4. If you just want your book available via the internet, Lulu is fine.
5. All the services subcontract their printing, and there are few incentives to produce a consistently high-quality product; if the physical quality of the book matters, you have to act like a real publisher, do your own printing contract, oversee the results, and so on.
Hope this is of some use!
Check out McAfee's free economics textbook: http://www.introecon.com/
Andrew, you may be interested in talking to Allen Hatcher. He's a math professor at Cornell who wrote a textbook, Algebraic Topology, and convinced the publisher (Cambridge Univ. Press) to let him post the whole thing as a free PDF on his website. A paperback copy costs $32 on Amazon.
This is a 550-page book which has quickly become one of the standard texts for an intro-level algebraic topology course. (I see that it's being used for the two-semester graduate sequence at Columbia.) Considering that Hatcher probably had to move heaven and earth to convince a publisher to accept this arrangement, he may be able to give good advice.
I'm not sure what an introductory statistics course at an American university would normally cover – would this cover more than is covered for Statistics A-Level in Britain? Statistics 1 for OCR Cambridge, Statistics 2 for OCR Cambridge and Statistics 3&4 for OCR Cambridge are three paperbacks by Steve Dobbs and Jane Miller which cover material tested by the Cambridge Board for A-Level; they cost about £10 apiece, so all three would currently come to about $45-$50. I like them a lot, but as I say I'm not sure what an American undergraduate would cover in an introductory course.
The real problem with a single expensive textbook, though, is that it means the student probably can't afford to go out and buy other books that haven't been assigned. But it's often the case that one will not really get the explanation of a topic in one book, while another makes sense – it's not that one book is better than the other on every single topic, one gets the hang of more topics if one can consult both. With three books one's chances are even better. It's really very important for the student to see this; it encourages the student to keep trying, looking for other ways into the subject, if that offered by the set text doesn't work.
Preston McAfee over at CalTech has a very comprehensive intro economics
textbook. He allows you to either download or order a printed version for, I believe, 11 dollars. It is a good book and covers many more topics than the typical intro text allowing for much flexibility in what to cover.
Students here sell their books back for half of the original price. But then the publisher comes up with a new, slightly modified edition. An unconcerned professor orders the newest edition, and bookstores will not buy back any used book. Go figure?!
I happily pay ~$150 for my text books.
Paying $2000 in-state public university tuition for a single 3-hour night class… Now that's robbery
I just stumbled upon this today:
http://www.mcafee.cc/Introecon/NSSP.html
It appears that some professors are offered a kickback to select certain books. Every university should make this against policy. Do universities have policies on this sort of thing generally?
-JD
Hi, a dozen previous people have been half right, it is the publishers who set prices and not authors. That being said, university bookstores often have exclusive contracts with those same publishers so it would need to be something you sold yourself or through the internet or if you live somewhere where the independent bookstore is still an option, then through a local non-chain. Even at Berkeley where the student bookstore defends its prices as raising money for student life, there's a pretty conspiracy-worthy amount of profit being made on each and every "Brand new edition! Completely revised and edited!!" freshman physics book that teaches angular momentum and the dipole moment just like nine earlier editions.
Good luck though, bust those cartels and see how it goes!
Kate.
@dan, it's my understanding that Cambridge U Press is rather good in this respect. Another example from them is David MacKay "Information Theory, Inference, and Learning Algorithms". Find the right publisher.
Check out the Jan/Feb issue of EDUCAUSE Review. (http://connect.educause.edu/er) The cover story is about the current state of the textbook market and the impact of new technologies. There are several related articles profiling textbook authors who have successfully published using a non-traditional approach.
Wow, $150 text book? That is scary.