Skip to content
 

Scholarpedia article on Bayesian Statistics

David Spiegelhalter and Ken Rice wrote this excellent short article on Bayesian statistics. I think it’s far superior to the Wikipedia articles on Bayes, most of which focus too much on discrete models for my taste.

7 Comments

  1. ZBicyclist says:

    I was unfamiliar with Scholarpedia, so poked around a bit. Links to Amazon.com seem to pop up a lot — is this venture sponsored?

    Example: ask for information about Weibull distributions, and you don't get an article but do get a link to an Amazon search.

    http://www.scholarpedia.org/article/Special:Searc

  2. Andrew Gelman says:

    I don't know. I just reviewed this particular article because they asked me to. That's all I know about it.

  3. Ken Rice says:

    Andrew – many thanks for the link, and the review. I agree the wikipedia emphasis is odd.

    ZB – Like a lot of other sites, they're getting some money for the direct Amazon ads which are on some pages, as well as the search links you mention. There's nothing on the Weibull because, so far, the scope of topics covered is limited – but will hopefully grow.

  4. It is indeed a really fine article.

    I particularly liked the way he handled the decision-theoretic problem of preparing for MRSA.

  5. gb says:

    So if the wikipedia article is flawed, why not improve it, and let a larger number of readers benefit from it?

  6. Andrew Gelman says:

    Gb: It takes work. There are a lot of different wikipedia articles on Bayes and they all have serious problems. If I were to just replace them with the Scholarpedia article, I'm afraid that people would get annoyed. But it would be a lot of work to rewrite each individual article. If anyone out there would like to do this, I agree that it would be a useful service.

  7. zbicyclist says:

    If you re-wrote the Wikipedia article, it might get un-re-written, and you might end up with one of those nasty revision / re-revision arguments I've heard about re Wikipedia.

    Plus, since the Scholarpedia articles have actual articles and have some level of review, they might "count" a bit in those counts academics have to worry about.

    In theory, Scholarpedia is a great idea and it will be interesting to see if it catches on. Having the pages look just like Wikipedia is good: It may be noted that utilization of an iconic schematic paradigm facilitates and optimizes mental assimilation. ;)