Where’s my biostatistician in shining armour?

Posted on April 10, 2008 by Cath@VWXYNot?

I’m going to try Richard’s approach here, even though it doesn’t seem to have generated much in the way of actual cloning help. Yet.
I need to brush up on my seriously rusty stats skills, which were never all that shiny to begin with. Can anyone recommend a good textbook that will remind me about power analysis and all that good stuff? Preferably something called “Statistics for Dummies who have Advanced Biology Degrees But Know Nothing About Statistics”.
Thanks!

About Cath@VWXYNot?

"one of the sillier science bloggers [...] I thought I should give a warning to the more staid members of the community." - Bob O'Hara, December 2010
This entry was posted in Uncategorized. Bookmark the permalink.

9 Responses to Where’s my biostatistician in shining armour?

  1. kate doc-in-training says:
    April 11, 2008 at 1:08 am

    I’d recommend Kirkwood’s Essential Medical statistics or D. Altman’s Practical Statistics for Medical Research. Kirkwood is easier to read, but Altman is more advanced and is often cited. Both are intro books though.

  2. Bob O'Hara says:
    April 11, 2008 at 6:08 am

    I’d suggest starting with The Cartoon Guide to Statistics.
    I’m not really sure what other books to suggest – it’s difficult for me to judge the quality of basic stats texts, but if you’re going to be doing quite a lot in the future, you could look at Gelman & Hill’s book.
    Quite frankly I’m shocked at a biologist wanting to know about power analysis. You’re meant to do the experiment first, and then come to the statistician and complain that you can’t work out how to analyse your data. It’s no fun otherwise.

  3. Richard P. Grant says:
    April 11, 2008 at 6:17 am

    Bob, this is Science 2.0. Cath wants to know what the results should be so she doesn’t have to do the experiment.

  4. Raf Aerts says:
    April 11, 2008 at 8:05 am

    In ecology the book by Siegel and Castellan is the bible of nonparametric statistics. It has a good table at the back, which enables you to pick the right test for any level of measurement/number of cases combination. It also helps you do aggregate your data down to 2×2 tables or one-sample cases if you realize you have been replicating your stuff on the wrong level.

  5. Bob O'Hara says:
    April 11, 2008 at 8:21 am

    I’ve got a copy of the 1956 version of the book Raf suggests. Evidently before Castellan came on board.
    I would imagine that a castellan wouldn’t be too please with people in shining armour wondering around the place knocking over the furniture.

  6. Neil Saunders says:
    April 11, 2008 at 9:05 am

    There’s a heap of excellent statistics resources on the Web. Many departments put their undergraduate statistics course material online, for example. This is where I always start when I need to learn something new or refresh my memory.
    Start with a well-phrased Google Search like statistics "power analysis" tutorial= and see where that takes you.
    Stat Pages – webpages that perform statistical calculations looks pretty good.
    When you find good stuff, stash it in your online social bookmarking system (del.icio.us, connotea) and you’ll have your own online, virtual textbook in no time.

  7. Cath Ennis says:
    April 11, 2008 at 4:23 pm

    Thanks all for the links and suggestions. I’ll head over to Amazon and look up some reviews. (That’s work related use of my computer, right? Even if I browse for CDs at the same time?). The cartoon book looks especially interesting!
    I don’t actually generate any data of my own, I just want to be able to verify that the stats in grant proposal texts I’m given are appropriate. At the moment I’m just having to trust that the PI / physician has done it right!

  8. Bob O'Hara says:
    April 13, 2008 at 10:09 am

    I just remembered something you might find useful – Statistics Notes in the British Medical Journal. Little essays explaining statistical topics to medics.

  9. Cath Ennis says:
    April 14, 2008 at 8:09 pm

    Thanks Bob, those look excellent!

Comments are closed.