Monday, March 5, 2012

The Book Visits Fisher

On their recent visit to St. Peter's Cathedral in Adelaide, colleagues Dan Navarro and Amy Perfors posed the book at the site of R. A. Fisher's remains. What is the probability of this happening if the null hypothesis were true?

Many thanks to Dan and Amy!
And see the book at Bayes' tomb, too.

12 comments:

  1. I'm going to call you out on this. Your book is EXTREMELY OVERPRICED. How do you actually expect Bayesian Statistics to catch on if only those with elite wallets can read it?? It's important to you to get paid, and support your lifestyle. So keeping it a secret keeps you getting gigs being THEE go to guru. When everybody starts to do it, you'll be less relevant and less paid. Really want to get on board, I know the absurdity of NHST first hand, but I'm not in a position to spend $$$$$$$$$ on a book. Glad the high price book help pay for you flight to Australia.

    ReplyDelete
  2. Dear Anonymous:

    Sorry that you're feeling that the price of the book is unjust. Perhaps some explanation would be helpful.

    I did not go to Australia. As the blog clearly states, the photos were kindly taken by colleagues who work there.

    The publisher sets the price of the book, and the publisher gets the vast majority of the money. (This is typical of all publishers, not just the publisher of this book.)

    The author gets only a small percentage of the proceeds from the book. I wrote the book out of sheer moral conviction that NHST needs to be replaced by Bayesian methods. I worked many thousands of hours across several years to create the book, with no advance royalties and no idea if anyone would actually buy it. The remuneration from the book has been FAR LESS than minimum wage would have paid. That’s the “lifestyle” to which you refer in your comment. I would have made much more money if I had stuck to pursuing research and not bothered writing the book at all.

    What is an appropriate price for a book such as this? Compared to many traditional (NHST) textbooks, the price is typical, with many being more expensive. As another comparison, consider the cost of one rock concert or broadway show. Many people are willing to pay the price for such a single-night event, but then balk at paying a similar price (or less) for a book that yields knowledge that lasts a lifetime.

    Solutions:
    1. Shop around for a discounted price. The author gets even less than the usual pittance.
    2. Buy a used copy, so my hard work earns zero of your hard-earned money.
    3. Start a wikibook so multiple authors contribute different sections free (without plagiarizing or pirating, of course).
    4. Start a publishing company that charges less for books but compensates authors better.

    ReplyDelete
  3. This exchange raises an interesting point. Is going with a traditional publisher (or traditional publishing contract) really the best way to disseminate academic work in book form? Some book authors seem to have negotiated deals that allow them to post the PDF of the book online for free or for a nominal charge and the book is available to buy in hard copy for those who prefer that format and are willing to pay the corresponding price. There are also ways to properly self-publish a book (with ISBN number and all that allow the book to be ordered by book stores, amazon.com etc.) using a print-on-demand publisher where almost all of the sale price goes to the author. Especially given that the role of the publisher in this particular book seems minimal (the author seems to have done almost everything himself including the cover design and promotion) it certainly is a shame (albeit typical for academic publishing) that the publisher gets almost all the profit ...

    ReplyDelete
  4. > What is the probability of this happening if the null hypothesis were true?

    I am not sure why, but this line struck me as absolutely hilarious.

    ReplyDelete
  5. It does seem somewhat perverse the small amount authors get paid given the work they put in.

    I would be interested to know if authors would be better off publishing through websites like OTexts . Although I must admit for text books I am willing to pay a premium to have a hard copy. That said, for an online book I would be happy to pay a suggested amount if I found the book useful.

    Coincidentally, one of the people behind OText, Rob J Hyndman, is referenced in DBDA.

    ReplyDelete
  6. It appears as though somebody is truly trying to make Fisher roll over in his grave. Hillarious!

    ReplyDelete
  7. Your book is excellent and I like your writing style. I bought two hard copies for my student and myself. Your work deserves more than the current price.

    ReplyDelete
  8. This comment has been removed by the author.

    ReplyDelete
  9. @John K. Kruschke, being from a developing country, I found the book too expensive as well and searched for an Indian Edition like mads and came to a posterior belief that Indian Edition of this book does not exist. Thanks for your answer to the Anonymous in March 2012, that I understand how this book proved to be more of a financial burden than an asset for you. Should you explore the possibility of an Indian edition with a prior of some extra royalty from there? This will also help people from Indian subcontinent to buy this amazing book at a price which is affordable to them as well as may help in getting you financial returns for your excellent effort. Definitely it will help your cause: NHST needs to be replaced by Bayesian methods.

    ReplyDelete
    Replies
    1. Thank you for your interest. International editions are created at the discretion of the publisher. It might help if you contacted the publisher directly to express your interest. Try the contact portal here: https://www.elsevier.com/support Thanks again.

      Delete
    2. Pleasantly surprised by your reply, thanks! I have written an email to the publisher requesting them to print an Indian edition, also provided them the stackexchange link where the book has highest votes for best introductory book for Bayesian Statistics (https://stats.stackexchange.com/questions/125/what-is-the-best-introductory-bayesian-statistics-textbook), though not very hopeful for a positive reply from them

      Delete